From nobody Mon Nov 3 15:42:45 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 851671A1A97; Mon, 3 Nov 2014 15:42:41 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.55 X-Spam-Level: X-Spam-Status: No, score=-1.55 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_DE=0.35] autolearn=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RUJb2K3CTZ3o; Mon, 3 Nov 2014 15:42:40 -0800 (PST) Received: from mailhost.informatik.uni-bremen.de (mailhost.informatik.uni-bremen.de [IPv6:2001:638:708:30c9::12]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CD7871A1A4A; Mon, 3 Nov 2014 15:42:39 -0800 (PST) X-Virus-Scanned: amavisd-new at informatik.uni-bremen.de Received: from smtp-fb3.informatik.uni-bremen.de (smtp-fb3.informatik.uni-bremen.de [134.102.224.120]) by mailhost.informatik.uni-bremen.de (8.14.5/8.14.5) with ESMTP id sA3Ngahr004743; Tue, 4 Nov 2014 00:42:36 +0100 (CET) Received: from alma.local.informatik.uni-bremen.de (robin.informatik.uni-bremen.de [134.102.218.1]) (using TLSv1.2 with cipher AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp-fb3.informatik.uni-bremen.de (Postfix) with ESMTPSA id 3C69495C; Tue, 4 Nov 2014 00:42:36 +0100 (CET) X-Mailer: emacs 25.0.50.1 (via feedmail 11-beta-1 I) From: Carsten Bormann To: bier@ietf.org, 6lo@ietf.org, roll@ietf.org Date: Tue, 04 Nov 2014 00:42:32 +0100 Message-ID: User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/25.0.50 (darwin) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Archived-At: http://mailarchive.ietf.org/arch/msg/bier/L_LgyBkStNDJHJD_-e0A6P5RNfI Cc: Andrzej.Duda@imag.fr Subject: [Bier] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 03 Nov 2014 23:42:42 -0000 In Hawaii, there will be a BOF about a multicast forwarding control scheme where a router on the way prefixes a multicast packet with a special header that tells downtree routers where to forward the multicast =E2=80=94 Bit Index Explicit Replication, or BIER. While BIER appears to be very focused on a carrier or datacenter perspective (see draft-kumar-bier-use-cases and the other draft-*-bier-*), there are applications in constrained node networks that would benefit from source-routed multicast. Several proposals on how to do source-routed multicast in constrained node networks are getting ready for consumption. With its current focus, I don=E2=80=99t think the BIER BOF itself is the ri= ght place to have a technical discussion about this. There is no ROLL meeting in Hawaii. The 6lo agenda is almost full. Should we do this in a side meeting? (If we do this before Tuesday, we can at least report in 6lo =E2=80=94 if we do it on Sunday, we might even h= ave an opinion to air during the BIER BOF.) Gr=C3=BC=C3=9Fe, Carsten From nobody Mon Nov 3 16:01:40 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2D7621A1AB7; Mon, 3 Nov 2014 16:01:38 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.999 X-Spam-Level: X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id M53E72a7P2LG; Mon, 3 Nov 2014 16:01:35 -0800 (PST) Received: from mail-wg0-x231.google.com (mail-wg0-x231.google.com [IPv6:2a00:1450:400c:c00::231]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3000D1A1AAB; Mon, 3 Nov 2014 16:01:35 -0800 (PST) Received: by mail-wg0-f49.google.com with SMTP id x13so13580770wgg.8 for ; Mon, 03 Nov 2014 16:01:33 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:reply-to:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=0XxPHITu3JqqJFxfPoWC28l7RbVABHljGwm1csY3iqs=; b=krA5KbxA5jXcjkc1D+QXwENwHbSp1Qqv+kqkqlf6ApmAp5kvcICtl+EneF3JEEn9dw wiAc5jk1uEuIEEdLnRKmZNv2agkgGpE5HP5s1rM4Ro1ngA1XC3IECzQT7Wy+oNOYLERn WJBKUhYowObTwcWzReI153lzfziCt2gUN9T6P29RachzDX2Pw/R4EZN2E3ijQl6aVFtH 5B2ofTx80yXbL3RJ6Q7XUNYckLVp/eMT//Zsxh7uu88g9eVCtWX973O5GlFScpFb+AAs +g+leNG8po0hEX5BwJvb0BZ54CNKdH6X2iFqpU+uru44DeqlfcrcDICiEIoPr0Wq//29 VRSQ== MIME-Version: 1.0 X-Received: by 10.180.106.104 with SMTP id gt8mr19600982wib.51.1415059293762; Mon, 03 Nov 2014 16:01:33 -0800 (PST) Received: by 10.180.20.79 with HTTP; Mon, 3 Nov 2014 16:01:33 -0800 (PST) In-Reply-To: References: Date: Mon, 3 Nov 2014 16:01:33 -0800 Message-ID: From: Greg Shepherd To: Carsten Bormann Content-Type: multipart/alternative; boundary=f46d04428ee28a78bc0506fd2a6b Archived-At: http://mailarchive.ietf.org/arch/msg/bier/HiuHBC6eNlOJ-mrLLqan4zJ2a1I Cc: Andrzej.Duda@imag.fr, roll@ietf.org, "bier@ietf.org" , 6lo@ietf.org Subject: Re: [Bier] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gjshep@gmail.com List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 04 Nov 2014 00:01:38 -0000 --f46d04428ee28a78bc0506fd2a6b Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Sunday works for me. Greg On Mon, Nov 3, 2014 at 3:42 PM, Carsten Bormann wrote: > In Hawaii, there will be a BOF about a multicast forwarding control > scheme where a router on the way prefixes a multicast packet with a > special header that tells downtree routers where to forward the > multicast =E2=80=94 Bit Index Explicit Replication, or BIER. > > While BIER appears to be very focused on a carrier or datacenter > perspective (see draft-kumar-bier-use-cases and the other > draft-*-bier-*), there are applications in constrained node networks > that would benefit from source-routed multicast. > > Several proposals on how to do source-routed multicast in constrained > node networks are getting ready for consumption. > > With its current focus, I don=E2=80=99t think the BIER BOF itself is the = right > place to have a technical discussion about this. > There is no ROLL meeting in Hawaii. > The 6lo agenda is almost full. > Should we do this in a side meeting? (If we do this before Tuesday, we > can at least report in 6lo =E2=80=94 if we do it on Sunday, we might even= have > an opinion to air during the BIER BOF.) > > Gr=C3=BC=C3=9Fe, Carsten > > _______________________________________________ > BIER mailing list > BIER@ietf.org > https://www.ietf.org/mailman/listinfo/bier > --f46d04428ee28a78bc0506fd2a6b Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Sunday works for me.

Greg

On Mon, Nov 3, 2014= at 3:42 PM, Carsten Bormann <cabo@tzi.org> wrote:
In Hawaii, there will be a BOF about a multicast forwar= ding control
scheme where a router on the way prefixes a multicast packet with a
special header that tells downtree routers where to forward the
multicast =E2=80=94 Bit Index Explicit Replication, or BIER.

While BIER appears to be very focused on a carrier or datacenter
perspective (see draft-kumar-bier-use-cases and the other
draft-*-bier-*), there are applications in constrained node networks
that would benefit from source-routed multicast.

Several proposals on how to do source-routed multicast in constrained
node networks are getting ready for consumption.

With its current focus, I don=E2=80=99t think the BIER BOF itself is the ri= ght
place to have a technical discussion about this.
There is no ROLL meeting in Hawaii.
The 6lo agenda is almost full.
Should we do this in a side meeting?=C2=A0 (If we do this before Tuesday, w= e
can at least report in 6lo =E2=80=94 if we do it on Sunday, we might even h= ave
an opinion to air during the BIER BOF.)

Gr=C3=BC=C3=9Fe, Carsten

_______________________________________________
BIER mailing list
BIER@ietf.org
ht= tps://www.ietf.org/mailman/listinfo/bier

--f46d04428ee28a78bc0506fd2a6b-- From nobody Tue Nov 4 07:37:11 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8CD9B1A8A3C; Tue, 4 Nov 2014 07:37:09 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -15.095 X-Spam-Level: X-Spam-Status: No, score=-15.095 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id FrcOBTTNdgM6; Tue, 4 Nov 2014 07:37:07 -0800 (PST) Received: from alln-iport-8.cisco.com (alln-iport-8.cisco.com [173.37.142.95]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 57CE91A8A3E; Tue, 4 Nov 2014 07:37:07 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1796; q=dns/txt; s=iport; t=1415115427; x=1416325027; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-id:content-transfer-encoding: mime-version; bh=wVYMahAxZyiJblGapVR4LjBVTEHGKNv5NnHtgPoVnF0=; b=Hh5FWkH7/nY7T0hTTSHrAjzVfSXaJH19Kuj5AOfs/I2bmBpOwG3BksMZ wGItilhyGbTWdMp7uQ+57txDBedxbABpCjbgcqKq54nkWFvVGO3rNIuPO 1I+ywo0+L2UJHqy9EAhCdS22Ou9L8cDq1WW201O75JRd42iO4n+dYtsZ+ s=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AhYFALjxWFStJA2L/2dsb2JhbABcgw5UWATOUwqHTQKBHxYBAQEBAX2EAgEBAQQBAQFkBwsMBAIBCBEDAQIBLiEGCxQJCAIEAQ0FiCwDEg3GaA2GKQEBAQEBAQEBAQEBAQEBAQEBAQEBARMEjlaCDysHBoRFAQSSIIlWghGPX4Zwg3hsgUiBAwEBAQ X-IronPort-AV: E=Sophos;i="5.07,313,1413244800"; d="scan'208";a="93236152" Received: from alln-core-6.cisco.com ([173.36.13.139]) by alln-iport-8.cisco.com with ESMTP; 04 Nov 2014 15:37:05 +0000 Received: from xhc-aln-x14.cisco.com (xhc-aln-x14.cisco.com [173.36.12.88]) by alln-core-6.cisco.com (8.14.5/8.14.5) with ESMTP id sA4Fb5kj021686 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 4 Nov 2014 15:37:05 GMT Received: from xmb-rcd-x06.cisco.com ([169.254.6.134]) by xhc-aln-x14.cisco.com ([173.36.12.88]) with mapi id 14.03.0195.001; Tue, 4 Nov 2014 09:37:05 -0600 From: "Rajiv Asati (rajiva)" To: "gjshep@gmail.com" , Carsten Bormann Thread-Topic: [Bier] Source-Routed Multicast Thread-Index: AQHP97/gpz3UHL2ClUOaJ+Zqtf7hK5xP+hGAgACgygA= Date: Tue, 4 Nov 2014 15:37:04 +0000 Message-ID: References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.3.140616 x-originating-ip: [10.82.221.67] Content-Type: text/plain; charset="Windows-1252" Content-ID: <476B0C17770E724C974B07F636D81E6F@emea.cisco.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Archived-At: http://mailarchive.ietf.org/arch/msg/bier/iv3M_joRlDsm2UzPEbiepvZL8Ts Cc: "Andrzej.Duda@imag.fr" , "roll@ietf.org" , "bier@ietf.org" , "6lo@ietf.org" <6lo@ietf.org> Subject: Re: [Bier] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 04 Nov 2014 15:37:09 -0000 +1 --=20 Cheers, Rajiv Asati Distinguished Engineer, Cisco -----Original Message----- From: "gjshep@gmail.com" Reply-To: "gjshep@gmail.com" Date: Monday, November 3, 2014 at 8:01 PM To: Carsten Bormann Cc: "Andrzej.Duda@imag.fr" , "roll@ietf.org" , "bier@ietf.org" , "6lo@ietf.org" <6lo@ietf.org> Subject: Re: [Bier] Source-Routed Multicast >Sunday works for me. > > >Greg > > >On Mon, Nov 3, 2014 at 3:42 PM, Carsten Bormann > wrote: > >In Hawaii, there will be a BOF about a multicast forwarding control >scheme where a router on the way prefixes a multicast packet with a >special header that tells downtree routers where to forward the >multicast =8B Bit Index Explicit Replication, or BIER. > >While BIER appears to be very focused on a carrier or datacenter >perspective (see draft-kumar-bier-use-cases and the other >draft-*-bier-*), there are applications in constrained node networks >that would benefit from source-routed multicast. > >Several proposals on how to do source-routed multicast in constrained >node networks are getting ready for consumption. > >With its current focus, I don=B9t think the BIER BOF itself is the right >place to have a technical discussion about this. >There is no ROLL meeting in Hawaii. >The 6lo agenda is almost full. >Should we do this in a side meeting? (If we do this before Tuesday, we >can at least report in 6lo =8B if we do it on Sunday, we might even have >an opinion to air during the BIER BOF.) > >Gr=FC=DFe, Carsten > >_______________________________________________ >BIER mailing list >BIER@ietf.org >https://www.ietf.org/mailman/listinfo/bier > > > > > From nobody Tue Nov 4 14:28:19 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 216BA1A0195; Tue, 4 Nov 2014 14:28:18 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -15.095 X-Spam-Level: X-Spam-Status: No, score=-15.095 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id CDy_dwAmPZBW; Tue, 4 Nov 2014 14:28:15 -0800 (PST) Received: from rcdn-iport-8.cisco.com (rcdn-iport-8.cisco.com [173.37.86.79]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6B5F81A0151; Tue, 4 Nov 2014 14:28:15 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=1697; q=dns/txt; s=iport; t=1415140096; x=1416349696; h=mime-version:subject:from:in-reply-to:date:cc: content-transfer-encoding:message-id:references:to; bh=9pcVRDWwKSx5jvh/X3Zn/4Qxs4+vSAFLwiDAtnTOX6E=; b=FW1xzN8H41ggRY+e9FNfRXfPGx0ZtmQNDC6dhE9G2NToTPsNRuGFcQPq uefJZ4dAg/BZcnOlfztkFE7/lhSOEf2wRtFZG0jsWGQjNy1s20oW70e1o cqt99mBPmUov/csAu08dqOQgY83lKSDjA7QjhIHv31NXVe3MEO42tcFHx w=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AhoLAAxSWVStJV2U/2dsb2JhbABbgw5UWbsKBQF0kloKh00CgSMWAQEBAQF9hAMBAQMBAQEBZAcLBQsLDjgnMAYTiDgJDcRJAQEBAQEBAQEBAQEBAQEBAQEBAQEBEwSGN4omMweDLYEeBYt7khOHdI5ehBkcL4JLAQEB X-IronPort-AV: E=Sophos;i="5.07,315,1413244800"; d="scan'208";a="369307074" Received: from rcdn-core-12.cisco.com ([173.37.93.148]) by rcdn-iport-8.cisco.com with ESMTP; 04 Nov 2014 22:28:15 +0000 Received: from [10.154.176.153] ([10.154.176.153]) by rcdn-core-12.cisco.com (8.14.5/8.14.5) with ESMTP id sA4MSDO4023842 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NO); Tue, 4 Nov 2014 22:28:14 GMT Content-Type: text/plain; charset=windows-1252 Mime-Version: 1.0 (Mac OS X Mail 7.3 \(1878.6\)) From: IJsbrand Wijnands In-Reply-To: Date: Tue, 4 Nov 2014 14:28:12 -0800 Content-Transfer-Encoding: quoted-printable Message-Id: References: To: Carsten Bormann X-Mailer: Apple Mail (2.1878.6) Archived-At: http://mailarchive.ietf.org/arch/msg/bier/5vIWcwo5OWxjmxzeP8ns_nhJ_4E Cc: Andrzej.Duda@imag.fr, roll@ietf.org, bier@ietf.org, 6lo@ietf.org Subject: Re: [Bier] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 04 Nov 2014 22:28:18 -0000 Hi Carsten, The ROLL case looks very interesting. Just one clarification, BIER is not Source-Routed multicast. i.e. we = don=92t include a list of hops that the packet needs to follow, we only = include final destination in the form of Bit Positions, and each router = along the path does a lookup on the Bit Position to determine where the = packet is to be sent. Thx, Ice. On 03 Nov 2014, at 15:42, Carsten Bormann wrote: > In Hawaii, there will be a BOF about a multicast forwarding control > scheme where a router on the way prefixes a multicast packet with a > special header that tells downtree routers where to forward the > multicast =97 Bit Index Explicit Replication, or BIER. >=20 > While BIER appears to be very focused on a carrier or datacenter > perspective (see draft-kumar-bier-use-cases and the other > draft-*-bier-*), there are applications in constrained node networks > that would benefit from source-routed multicast. >=20 > Several proposals on how to do source-routed multicast in constrained > node networks are getting ready for consumption. >=20 > With its current focus, I don=92t think the BIER BOF itself is the = right > place to have a technical discussion about this. > There is no ROLL meeting in Hawaii. > The 6lo agenda is almost full. > Should we do this in a side meeting? (If we do this before Tuesday, = we > can at least report in 6lo =97 if we do it on Sunday, we might even = have > an opinion to air during the BIER BOF.) >=20 > Gr=FC=DFe, Carsten >=20 > _______________________________________________ > BIER mailing list > BIER@ietf.org > https://www.ietf.org/mailman/listinfo/bier From nobody Tue Nov 4 14:52:13 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 997461A02F1; Tue, 4 Nov 2014 14:52:03 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.85 X-Spam-Level: X-Spam-Status: No, score=-3.85 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_DE=0.35, RCVD_IN_DNSWL_MED=-2.3] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id PCsCdCXRighI; Tue, 4 Nov 2014 14:52:00 -0800 (PST) Received: from mailhost.informatik.uni-bremen.de (mailhost.informatik.uni-bremen.de [134.102.201.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 233A01A0277; Tue, 4 Nov 2014 14:51:59 -0800 (PST) X-Virus-Scanned: amavisd-new at informatik.uni-bremen.de Received: from smtp-fb3.informatik.uni-bremen.de (smtp-fb3.informatik.uni-bremen.de [134.102.224.120]) by mailhost.informatik.uni-bremen.de (8.14.5/8.14.5) with ESMTP id sA4MosoV029060; Tue, 4 Nov 2014 23:50:54 +0100 (CET) Received: from [192.168.217.113] (p54892E83.dip0.t-ipconnect.de [84.137.46.131]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by smtp-fb3.informatik.uni-bremen.de (Postfix) with ESMTPSA id 73A252FE; Tue, 4 Nov 2014 23:50:52 +0100 (CET) Content-Type: text/plain; charset=windows-1252 Mime-Version: 1.0 (Mac OS X Mail 8.0 \(1990.1\)) From: Carsten Bormann In-Reply-To: Date: Tue, 4 Nov 2014 23:50:50 +0100 X-Mao-Original-Outgoing-Id: 436834250.592611-9c57e15d318f9248d6425265b403cca7 Content-Transfer-Encoding: quoted-printable Message-Id: <300194F7-D81F-4F2B-923A-BA6EF09DC334@tzi.org> References: To: IJsbrand Wijnands X-Mailer: Apple Mail (2.1990.1) Archived-At: http://mailarchive.ietf.org/arch/msg/bier/jxdCdjvpbeW0JEMuAUK2B3Agn58 Cc: Andrzej.Duda@imag.fr, roll@ietf.org, bier@ietf.org, 6lo@ietf.org Subject: Re: [Bier] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 04 Nov 2014 22:52:03 -0000 > Just one clarification, BIER is not Source-Routed multicast. i.e. we = don=92t include a list of hops that the packet needs to follow, we only = include final destination in the form of Bit Positions, and each router = along the path does a lookup on the Bit Position to determine where the = packet is to be sent. Indeed, that=92s why I titled this E-mail =93Source-Routed Multicast=94 = so it is clear it is not the same as BIER. Source-routing is important = for RPL=92s non-storing mode, where intermediate routers just don=92t = know what is below them. (BIER could be applied to RPL=92s storing = mode, but then that already has a multicast approach written up, and = it=92s also not that clear that storing mode actually exists. The next = question would then be whether RPL networks are stable enough to have a = stable egress router numbering applied to them.) It seems several people would be able to make a Sunday hallway chat = about this general approach. Since I=92m officially at a couple of = meetings (in parallel) from 1600, how about 1500? Gr=FC=DFe, Carsten From nobody Wed Nov 5 12:46:37 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9CF281A1BCE; Mon, 3 Nov 2014 18:18:19 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.495 X-Spam-Level: X-Spam-Status: No, score=-2.495 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5B71E_waDL5p; Mon, 3 Nov 2014 18:18:18 -0800 (PST) Received: from tuna.sandelman.ca (tuna.sandelman.ca [IPv6:2607:f0b0:f:3:216:3eff:fe7c:d1f3]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2CD291A1BBD; Mon, 3 Nov 2014 18:18:18 -0800 (PST) Received: from sandelman.ca (obiwan.sandelman.ca [IPv6:2607:f0b0:f:2::247]) by tuna.sandelman.ca (Postfix) with ESMTP id 9A95920028; Mon, 3 Nov 2014 21:19:54 -0500 (EST) Received: by sandelman.ca (Postfix, from userid 179) id 12933637F4; Mon, 3 Nov 2014 21:18:16 -0500 (EST) Received: from sandelman.ca (localhost [127.0.0.1]) by sandelman.ca (Postfix) with ESMTP id F1BA263740; Mon, 3 Nov 2014 21:18:16 -0500 (EST) From: Michael Richardson To: Carsten Bormann In-Reply-To: References: X-Mailer: MH-E 8.2; nmh 1.3-dev; GNU Emacs 23.4.1 X-Face: $\n1pF)h^`}$H>Hk{L"x@)JS7<%Az}5RyS@k9X%29-lHB$Ti.V>2bi.~ehC0; <'$9xN5Ub# z!G,p`nR&p7Fz@^UXIn156S8.~^@MJ*mMsD7=QFeq%AL4m Sender: mcr@sandelman.ca Archived-At: http://mailarchive.ietf.org/arch/msg/bier/lDw6k7JGPhKhNykBR4EH8tsYt28 X-Mailman-Approved-At: Wed, 05 Nov 2014 12:46:34 -0800 Cc: Andrzej.Duda@imag.fr, roll@ietf.org, bier@ietf.org, 6lo@ietf.org Subject: Re: [Bier] [6lo] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 04 Nov 2014 02:18:19 -0000 --=-=-= Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Carsten Bormann wrote: > In Hawaii, there will be a BOF about a multicast forwarding control > scheme where a router on the way prefixes a multicast packet with a > special header that tells downtree routers where to forward the > multicast =E2=80=94 Bit Index Explicit Replication, or BIER. > While BIER appears to be very focused on a carrier or datacenter > perspective (see draft-kumar-bier-use-cases and the other > draft-*-bier-*), there are applications in constrained node networks > that would benefit from source-routed multicast. I take your point. > With its current focus, I don=E2=80=99t think the BIER BOF itself is = the right > place to have a technical discussion about this. On the contrary, I think it is important at a BOF for this topic to be brought up, if ONLY for it to be CLEARLY ruled out of scope. > There is no ROLL meeting in Hawaii. > The 6lo agenda is almost full. ROLL would have to be rechartered to work on it; this is not to say that th= is could not happen, but rather to say that even if we were meeting, discussing it might be out of scope anyway. Having said that, this thread is most welcome. > Should we do this in a side meeting? (If we do this before Tuesday, = we > can at least report in 6lo =E2=80=94 if we do it on Sunday, we might = even have > an opinion to air during the BIER BOF.) I abhor "side meetings" --- either have a proper BOF in a room, or get a bunch of beers is my opinion. =2D- Michael Richardson , Sandelman Software Works IETF ROLL WG co-chair. http://datatracker.ietf.org/wg/roll/charter/ --=-=-= Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQEVAwUBVFg3ZYCLcPvd0N1lAQIMlgf9EKuKoUzuLFaam9hhusy+J7IiNSiMxtnV tCTOl2n/Nd6Ba+lPV7f4U2TzsebUWGBemGkb5//NF4JqfNtVNYJhObtB3MHiNnuE SldbYiCin4dD4eWPdSseCrTVXd36wnGPtwdgrXGuzQKlfgCiDe/kOYCrxzmTkW4K vvQsRwf2pw5rVb2E0elmm0KQ3NyO4nB/vaH/7EZ9tS0tz6TGj9VLvvE/iiVmT+Jx Zbshdb+Csj/jmOqc3EQzAD2wJrmIgXaWI4ML9EblV/LrykXO5CK1Srk82JLsDfE8 gcAj4FzL9cY6beJnBPqaFBw73MPqrWqIAyF7fdns+c5y1qFm8hdErw== =O2d8 -----END PGP SIGNATURE----- --=-=-=-- From nobody Wed Nov 5 12:46:39 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 157391A1A25; Tue, 4 Nov 2014 18:10:52 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -15.095 X-Spam-Level: X-Spam-Status: No, score=-15.095 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id q4LwR_V74l7S; Tue, 4 Nov 2014 18:10:50 -0800 (PST) Received: from alln-iport-8.cisco.com (alln-iport-8.cisco.com [173.37.142.95]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0F8CC1A8779; Tue, 4 Nov 2014 18:10:50 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=2898; q=dns/txt; s=iport; t=1415153450; x=1416363050; h=from:to:cc:subject:date:message-id:references: in-reply-to:content-transfer-encoding:mime-version; bh=ESeaVLb5V2p1wYAUTyV5PoiAaAilDeGrMji6dhkUaWE=; b=YfY7kMqIQ9ge8f/ffm2c02HqmrMW5/J1iGrnSLd8xLXPnFxZyXpxzUtb /ZT58DcJbDaMr02N55oZNPvRx8pEAk2q5Qbb3Md1/E9oB3uznHJirFFM2 liwe6p1dDUcncGfKLwubuBdyHCC0RC79XrlPA37/DBLVdY4GBh9ORe0AE 8=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AhgFAMWGWVStJV2Z/2dsb2JhbABbgw5UWQTOYgqHTQKBGBYBAQEBAX2EAgEBAQQBAQFkBwsMBAIBCBEEAQEBCh0HJwsUCQgCBAENBQiIOQ3GfwEBAQEBAQEBAQEBAQEBAQEBAQEBARMEkDgnMQcGgyeBHgWLPYZkoj+DeGyBBkKBAwEBAQ X-IronPort-AV: E=Sophos;i="5.07,316,1413244800"; d="scan'208";a="93427750" Received: from rcdn-core-2.cisco.com ([173.37.93.153]) by alln-iport-8.cisco.com with ESMTP; 05 Nov 2014 02:10:49 +0000 Received: from xhc-rcd-x08.cisco.com (xhc-rcd-x08.cisco.com [173.37.183.82]) by rcdn-core-2.cisco.com (8.14.5/8.14.5) with ESMTP id sA52AnOg031183 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 5 Nov 2014 02:10:49 GMT Received: from xmb-rcd-x01.cisco.com ([169.254.1.207]) by xhc-rcd-x08.cisco.com ([173.37.183.82]) with mapi id 14.03.0195.001; Tue, 4 Nov 2014 20:10:49 -0600 From: "Pascal Thubert (pthubert)" To: Routing Over Low power and Lossy networks , Carsten Bormann Thread-Topic: [Roll] [Bier] Source-Routed Multicast Thread-Index: AQHP+Jc6eJPa1iU6+EqbNCABEErj65xRRsVA Date: Wed, 5 Nov 2014 02:10:47 +0000 Deferred-Delivery: Wed, 5 Nov 2014 02:10:00 +0000 Message-ID: References: In-Reply-To: Accept-Language: fr-FR, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.154.177.243] Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Archived-At: http://mailarchive.ietf.org/arch/msg/bier/iZCwChuGf02lg8lvlRbgVR3mWvc X-Mailman-Approved-At: Wed, 05 Nov 2014 12:46:34 -0800 Cc: "Andrzej.Duda@imag.fr" , "bier@ietf.org" , "6lo@ietf.org" <6lo@ietf.org> Subject: Re: [Bier] [Roll] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 05 Nov 2014 02:10:52 -0000 State in BIER (one bitmap per child) is so limited and controlled that a c= onstrained node that cannot do storing may be able to still do BIER. For su= ch devices, we could propose a new mode of operation for RPL that is BIER o= nly, for both unicast and multicast traffic. If the preferred parent tree is acceptable for downward traffic, the DAOs w= ould percolate up by ORing the bitmaps of the children and passing that to = the preferred parent. Decision to forward down to a child (or of a child to= process an incoming message) would be a simple AND of the bitmap in the pa= cket and that of the child, regardless of unicast or multicast. Cheers, Pascal > -----Original Message----- > From: Roll [mailto:roll-bounces@ietf.org] On Behalf Of IJsbrand Wijnands > (iwijnand) > Sent: mardi 4 novembre 2014 14:28 > To: Carsten Bormann > Cc: Andrzej.Duda@imag.fr; roll@ietf.org; bier@ietf.org; 6lo@ietf.org > Subject: Re: [Roll] [Bier] Source-Routed Multicast >=20 > Hi Carsten, >=20 > The ROLL case looks very interesting. >=20 > Just one clarification, BIER is not Source-Routed multicast. i.e. we don'= t include > a list of hops that the packet needs to follow, we only include final des= tination in > the form of Bit Positions, and each router along the path does a lookup o= n the > Bit Position to determine where the packet is to be sent. >=20 > Thx, >=20 > Ice. >=20 > On 03 Nov 2014, at 15:42, Carsten Bormann wrote: >=20 > > In Hawaii, there will be a BOF about a multicast forwarding control > > scheme where a router on the way prefixes a multicast packet with a > > special header that tells downtree routers where to forward the > > multicast - Bit Index Explicit Replication, or BIER. > > > > While BIER appears to be very focused on a carrier or datacenter > > perspective (see draft-kumar-bier-use-cases and the other > > draft-*-bier-*), there are applications in constrained node networks > > that would benefit from source-routed multicast. > > > > Several proposals on how to do source-routed multicast in constrained > > node networks are getting ready for consumption. > > > > With its current focus, I don't think the BIER BOF itself is the right > > place to have a technical discussion about this. > > There is no ROLL meeting in Hawaii. > > The 6lo agenda is almost full. > > Should we do this in a side meeting? (If we do this before Tuesday, > > we can at least report in 6lo - if we do it on Sunday, we might even > > have an opinion to air during the BIER BOF.) > > > > Gr=FC=DFe, Carsten > > > > _______________________________________________ > > BIER mailing list > > BIER@ietf.org > > https://www.ietf.org/mailman/listinfo/bier >=20 > _______________________________________________ > Roll mailing list > Roll@ietf.org > https://www.ietf.org/mailman/listinfo/roll From nobody Thu Nov 6 13:52:32 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3BE361A90CD for ; Thu, 6 Nov 2014 13:52:28 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.6 X-Spam-Level: X-Spam-Status: No, score=-2.6 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, J_BACKHAIR_16=1, J_CHICKENPOX_12=0.6, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cYhxcHkd27YH for ; Thu, 6 Nov 2014 13:52:18 -0800 (PST) Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 076951A90CB for ; Thu, 6 Nov 2014 13:52:17 -0800 (PST) X-AuditID: c6180641-f79916d00000623a-e9-545b92b92b61 Received: from EUSAAHC004.ericsson.se (Unknown_Domain [147.117.188.84]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id 49.01.25146.AB29B545; Thu, 6 Nov 2014 16:24:42 +0100 (CET) Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC004.ericsson.se ([147.117.188.84]) with mapi id 14.03.0174.001; Thu, 6 Nov 2014 16:52:16 -0500 From: Albert Tian To: "bier@ietf.org" Thread-Topic: blocked based encoding and list based encoding for BIER Thread-Index: AQHP+gvtdAEkqA2KAkWbuRi5DQw54g== Date: Thu, 6 Nov 2014 21:52:14 +0000 Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.4.140807 x-originating-ip: [147.117.188.9] Content-Type: multipart/mixed; boundary="_005_D0812D8D311743AlbertTianericssoncom_" MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFvrOIsWRmVeSWpSXmKPExsUyuXRPiO6uSdEhBgfOSlssnbGHyYHRY8mS n0wBjFFcNimpOZllqUX6dglcGfsWLGcp+LGXqWLCtAWMDYyTNzF1MXJySAiYSHx+v5ERwhaT uHBvPVsXIxeHkMARRon1u7+zQDjLGCW+LFjJBlLFJqAj8Wr2A7AOEQFlifOHGtlBipgFVjFK 9G34xgKSEBZwkFi5+jlUkavE7G07oGw9iYe7ZoINYhFQkehafxzM5hUwk7j2YS/YSYxAZ3w/ tQbMZhYQl7j1ZD7UqSISDy+eZoOwRSVePv7HCmKLAs18tmEzO0RcUWJf/3R2iN5QiTer5kPN F5Q4OfMJywRGkVlIxs5CUjYLSRlEPFpi4r9fULaOxILdn9ggbG2JZQtfM8PYZw48hppjLdHZ +IUZU42dxNStK6FqjCTWzVkFZIPCazejxPSbuxhhmruOTYIqUpSY0v2QHcK2kTj/aCZQDQeQ bSDx70M8RO9ORonFsz4xw/QuX/iMHVnvAkbhVYwcpcWpZbnpRoabGIEp5ZgEm+MOxgWfLA8x CnAwKvHwbiiOChFiTSwrrsw9xCjNwaIkzqtZPS9YSCA9sSQ1OzW1ILUovqg0J7X4ECMTB6dU A2Py8ncl7xyqD+R0uJ7f1fL4nJ7Deevchu/FgrsWJCv8ZvYorTiTdDFZp7t6kweXcoTr1F1f 2BcGMc/8rt6aGWs33fhxx5ZFG4q/CukwGn5x+e7jPm+zkfUpc4EHr95HnNRI6NPfKve65fJO E9MPOzhZ3zcuzVhwIyH28ursF3VbdS5o1vtEH1BiKc5INNRiLipOBAAnkzTgCgMAAA== Archived-At: http://mailarchive.ietf.org/arch/msg/bier/qf6_5uevFpsrI_0jkumWZi-zoOo Cc: Antoni Przygienda , Ramanathan Lakshmikanthan , Jeff Tantsura Subject: [Bier] blocked based encoding and list based encoding for BIER X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 06 Nov 2014 21:52:28 -0000 --_005_D0812D8D311743AlbertTianericssoncom_ Content-Type: multipart/alternative; boundary="_000_D0812D8D311743AlbertTianericssoncom_" --_000_D0812D8D311743AlbertTianericssoncom_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi, We are looking at some possible improvements in the BIER encoding to better= handle sparsely populated multicast groups in large networks. In these cas= es, the bigStrings would have very few ones in between many zeros. We have two ideas, as described in the following two short drafts attached = in this email. We would upload the drafts as soon as the tool opens. 1. Blocked based encoding Essentially we divide the BitString into several blocks, one for each regio= n in the network. We would only include the regions where we have receivers= . 2. List based encoding Here we directly encode the list of BFR-ids of the receivers in place of th= e BitString. The associated packet processing is also included in the drafts. We would like to socialize the ideas among the bier drinkers. Your feedback= is really appreciated. Cheers, Albert --_000_D0812D8D311743AlbertTianericssoncom_ Content-Type: text/html; charset="us-ascii" Content-ID: <3AA381E68A5E744B8D1E0EB822508FF5@ericsson.com> Content-Transfer-Encoding: quoted-printable
Hi,

We are looking at some possible improvements in the BIER encoding to b= etter handle sparsely populated multicast groups in large networks. In thes= e cases, the bigStrings would have very few ones in between many zeros.

We have two ideas, as described in the following two short drafts atta= ched in this email. We would upload the drafts as soon as the tool opens.

1. Blocked based encoding

Essentially we divide the BitString into several blocks, one for each = region in the network. We would only include the regions where we have rece= ivers.

2. List based encoding

Here we directly encode the list of BFR-ids of the receivers in place = of the BitString.

The associated packet processing is also included in the drafts.

We would like to socialize the ideas among the bier drinkers. Your fee= dback is really appreciated.

Cheers,
Albert

--_000_D0812D8D311743AlbertTianericssoncom_-- --_005_D0812D8D311743AlbertTianericssoncom_ Content-Type: text/plain; name="draft-tian-bier-blocks-00.txt" Content-Description: draft-tian-bier-blocks-00.txt Content-Disposition: attachment; filename="draft-tian-bier-blocks-00.txt"; size=15737; creation-date="Thu, 06 Nov 2014 21:52:14 GMT"; modification-date="Thu, 06 Nov 2014 21:52:14 GMT" Content-ID: <94FDE6801AA4144288F2B2BF5D8F77B2@ericsson.com> Content-Transfer-Encoding: base64 DQoNCg0KDQpXb3JraW5nIEdyb3VwICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgIEEuIFRpYW4NCkludGVybmV0LURyYWZ0ICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgUi4gTGFrc2htaQ0KSW50ZW5kZWQgc3RhdHVz OiBTdGFuZGFyZHMgVHJhY2sgICAgICAgICAgICAgICAgICAgICAgICAgICAgIEouIFRhbnRzdXJh DQpFeHBpcmVzOiBNYXkgOSwgMjAxNSAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgIEEuIFByenlnaWVuZGENCiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICBFcmljc3NvbiwgSW5jLg0KICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICBOb3ZlbWJlciA1LCAyMDE0DQoNCg0K ICAgICAgICAgICAgICAgICAgICAgQmxvY2sgQmFzZWQgRW5jb2RpbmcgZm9yIEJJRVINCiAgICAg ICAgICAgICAgICAgICAgICAgZHJhZnQtdGlhbi1iaWVyLWJsb2Nrcy0wMA0KDQpBYnN0cmFjdA0K DQogICBUaGlzIGRvY3VtZW50IGRlc2NyaWJlcyBhIG5ldyBibG9ja2VkIGJhc2VkIGVuY29kaW5n IHNjaGVtZSBmb3IgQml0DQogICBJbmRleCBFeHBsaWNpdCBSZXBsaWNhdGlvbiAoQklFUikgbXVs dGljYXN0IGZvcndhcmRpbmcgc29sdXRpb24gdG8NCiAgIGJldHRlciBoYW5kbGUgc3BhcnNlbHkg cG9wdWxhdGVkIG11bHRpY2FzdCBncm91cHMgaW4gbGFyZ2UgbmV0d29ya3MuDQogICBUaGUgYXNz b2NpYXRlZCBwYWNrZXQgcHJvY2Vzc2luZyBpcyBhbHNvIGRlc2NyaWJlZC4NCg0KU3RhdHVzIG9m IFRoaXMgTWVtbw0KDQogICBUaGlzIEludGVybmV0LURyYWZ0IGlzIHN1Ym1pdHRlZCBpbiBmdWxs IGNvbmZvcm1hbmNlIHdpdGggdGhlDQogICBwcm92aXNpb25zIG9mIEJDUCA3OCBhbmQgQkNQIDc5 Lg0KDQogICBJbnRlcm5ldC1EcmFmdHMgYXJlIHdvcmtpbmcgZG9jdW1lbnRzIG9mIHRoZSBJbnRl cm5ldCBFbmdpbmVlcmluZw0KICAgVGFzayBGb3JjZSAoSUVURikuICBOb3RlIHRoYXQgb3RoZXIg Z3JvdXBzIG1heSBhbHNvIGRpc3RyaWJ1dGUNCiAgIHdvcmtpbmcgZG9jdW1lbnRzIGFzIEludGVy bmV0LURyYWZ0cy4gIFRoZSBsaXN0IG9mIGN1cnJlbnQgSW50ZXJuZXQtDQogICBEcmFmdHMgaXMg YXQgaHR0cDovL2RhdGF0cmFja2VyLmlldGYub3JnL2RyYWZ0cy9jdXJyZW50Ly4NCg0KICAgSW50 ZXJuZXQtRHJhZnRzIGFyZSBkcmFmdCBkb2N1bWVudHMgdmFsaWQgZm9yIGEgbWF4aW11bSBvZiBz aXggbW9udGhzDQogICBhbmQgbWF5IGJlIHVwZGF0ZWQsIHJlcGxhY2VkLCBvciBvYnNvbGV0ZWQg Ynkgb3RoZXIgZG9jdW1lbnRzIGF0IGFueQ0KICAgdGltZS4gIEl0IGlzIGluYXBwcm9wcmlhdGUg dG8gdXNlIEludGVybmV0LURyYWZ0cyBhcyByZWZlcmVuY2UNCiAgIG1hdGVyaWFsIG9yIHRvIGNp dGUgdGhlbSBvdGhlciB0aGFuIGFzICJ3b3JrIGluIHByb2dyZXNzLiINCg0KICAgVGhpcyBJbnRl cm5ldC1EcmFmdCB3aWxsIGV4cGlyZSBvbiBNYXkgOSwgMjAxNS4NCg0KQ29weXJpZ2h0IE5vdGlj ZQ0KDQogICBDb3B5cmlnaHQgKGMpIDIwMTQgSUVURiBUcnVzdCBhbmQgdGhlIHBlcnNvbnMgaWRl bnRpZmllZCBhcyB0aGUNCiAgIGRvY3VtZW50IGF1dGhvcnMuICBBbGwgcmlnaHRzIHJlc2VydmVk Lg0KDQogICBUaGlzIGRvY3VtZW50IGlzIHN1YmplY3QgdG8gQkNQIDc4IGFuZCB0aGUgSUVURiBU cnVzdCdzIExlZ2FsDQogICBQcm92aXNpb25zIFJlbGF0aW5nIHRvIElFVEYgRG9jdW1lbnRzDQog ICAoaHR0cDovL3RydXN0ZWUuaWV0Zi5vcmcvbGljZW5zZS1pbmZvKSBpbiBlZmZlY3Qgb24gdGhl IGRhdGUgb2YNCiAgIHB1YmxpY2F0aW9uIG9mIHRoaXMgZG9jdW1lbnQuICBQbGVhc2UgcmV2aWV3 IHRoZXNlIGRvY3VtZW50cw0KICAgY2FyZWZ1bGx5LCBhcyB0aGV5IGRlc2NyaWJlIHlvdXIgcmln aHRzIGFuZCByZXN0cmljdGlvbnMgd2l0aCByZXNwZWN0DQogICB0byB0aGlzIGRvY3VtZW50LiAg Q29kZSBDb21wb25lbnRzIGV4dHJhY3RlZCBmcm9tIHRoaXMgZG9jdW1lbnQgbXVzdA0KICAgaW5j bHVkZSBTaW1wbGlmaWVkIEJTRCBMaWNlbnNlIHRleHQgYXMgZGVzY3JpYmVkIGluIFNlY3Rpb24g NC5lIG9mDQoNCg0KDQoNClRpYW4sIGV0IGFsLiAgICAgICAgICAgICAgIEV4cGlyZXMgTWF5IDks IDIwMTUgICAgICAgICAgICAgICAgICBbUGFnZSAxXQ0KDA0KSW50ZXJuZXQtRHJhZnQgICAgICAg IEJsb2NrIEJhc2VkIEVuY29kaW5nIGZvciBCSUVSICAgICAgICBOb3ZlbWJlciAyMDE0DQoNCg0K ICAgdGhlIFRydXN0IExlZ2FsIFByb3Zpc2lvbnMgYW5kIGFyZSBwcm92aWRlZCB3aXRob3V0IHdh cnJhbnR5IGFzDQogICBkZXNjcmliZWQgaW4gdGhlIFNpbXBsaWZpZWQgQlNEIExpY2Vuc2UuDQoN ClRhYmxlIG9mIENvbnRlbnRzDQoNCiAgIDEuICBJbnRyb2R1Y3Rpb24gIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuICAgMg0KICAgICAxLjEuICBSZXF1aXJl bWVudHMgTGFuZ3VhZ2UgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gICAzDQog ICAgIDEuMi4gIEFjcm9ueW1zICAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gLiAuIC4gLiAgIDMNCiAgIDIuICBQcm9wb3NlZCBTb2x1dGlvbiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuICAgMw0KICAgICAyLjEuICBOZXcgRW5jb2Rpbmcg IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gICAzDQogICAgICAg Mi4xLjEuICBFeGFtcGxlIE9uZSAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gLiAgIDMNCiAgICAgICAyLjEuMi4gIEV4YW1wbGUgVHdvIC4gLiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuICAgNA0KICAgICAyLjIuICBQYWNrZXQgUHJvY2Vzc2luZyBG bG93ICAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gICA1DQogICAzLiAgQ29uY2x1 c2lvbnMgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAg IDcNCiAgIDQuICBJQU5BIENvbnNpZGVyYXRpb25zIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuICAgNw0KICAgNS4gIFNlY3VyaXR5IENvbnNpZGVyYXRpb25zIC4gLiAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gICA3DQogICA2LiAgUmVmZXJlbmNlcyAg LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAgIDcNCiAg ICAgNi4xLiAgTm9ybWF0aXZlIFJlZmVyZW5jZXMgIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuICAgNw0KICAgICA2LjIuICBJbmZvcm1hdGl2ZSBSZWZlcmVuY2VzICAuIC4gLiAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gICA3DQogICBBdXRob3JzJyBBZGRyZXNzZXMgIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAgIDcNCg0KMS4gIElu dHJvZHVjdGlvbg0KDQogICBCaXQgSW5kZXggRXhwbGljaXQgUmVwbGljYXRpb24gKEJJRVIpLCBh cyBkZWZpbmVkIGluIFtJLUQuZHJhZnQtDQogICB3aWpuYW5kcy1iaWVyLWFyY2hpdGVjdHVyZV0g aXMgYSBuZXcgYXJjaGl0ZWN0dXJlIGZvciBmb3J3YXJkaW5nDQogICBtdWx0aWNhc3QgZGF0YSBw YWNrZXRzLCBieSBlbmNvZGluZyB0aGUgbXVsdGljYXN0IGdyb3VwIG1lbWJlcnNoaXANCiAgIGRp cmVjdGx5IGluIGEgbmV3IHBhY2tldCBoZWFkZXIgdXNpbmcgYSBiaXRTdHJpbmcuICBCSUVSIGFz c2lnbmVzIGFuDQogICB1bmlxdWUgMTYtYml0IElEIChCRlItaWQpIHRvIGVhY2ggQml0IEZvcndh cmRpbmcgUm91dGVyIChCRlIpLA0KICAgcm91dGVycyB0aGF0IHN1cHBvcnRzIEJJRVIuICBCRlIt aWQgY29udGFpbnMgaGlnaGVyIDgtYml0IGFzIFNJDQogICAod2hpY2ggY2FuIGJlIHVzZWQgdG8g cmVwcmVzZW50IGEgcmVnaW9uKSwgYW5kIGxvd2VyIDgtYml0IGFzIGEgYml0DQogICBwb3NpdGlv biBpbnRvIGEgMjU2LWJpdCBiaXRTdHJpbmcuICBFYWNoIHNldCBiaXQgaW4gdGhlIGJpdFN0cmlu Zw0KICAgcmVwcmVzZW50IGEgbWVtYmVyIGluIHRoZSBncm91cCBmb3Igd2hpY2ggdGhlIHBhY2tl dCBuZWVkcyB0byBiZQ0KICAgZGVsaXZlcmVkLiAgQklFUiB3b3JrcyB3ZWxsIHdoZW4gdGhlIG5l dHdvcmsgaXMgbm90IHRvbyBiaWcgKGxlc3MNCiAgIHRoYW4gMjU2IEJGUnMpLiAgV2hlbiB0aGUg bmV0d29yayBoYXMgbW9yZSB0aGFuIDI1NiBCRlJzLCBtdWx0aXBsZQ0KICAgU0lzIHdpbGwgYmUg cmVxdWlyZWQuICBJdCBpcyBsaWtlbHkgdGhhdCB0aGUgbWVtYmVycyBvZiBhIGdyb3VwIGNhbg0K ICAgc3BhbiBhY3Jvc3MgbXVsdGlwbGUgcmVnaW9ucyBhbmQgdGhlcmVmb3JlIGhhdmUgZGlmZmVy ZW50IFNJcy4gIEluDQogICB0aGlzIGNhc2UsIHJlcGxpY2F0aW9uIGF0IHRoZSBpbmdyZXNzIHdp bGwgYmUgdXNlZCwgb25lIHBhY2tldCBmb3INCiAgIGVhY2ggU0kgKHJlZ2lvbikuDQoNCiAgIElu IHJlYWxpdHksIG1vc3Qgb2YgdGhlIG11bHRpY2FzdCBncm91cHMgYXJlIHNwYXJzZSwgd2hlbiBj b21wYXJlZCB0bw0KICAgdGhlIHRvdGFsIG51bWJlciBvZiByb3V0ZXJzIGluIGEgbmV0d29yay4g IFVzdWFsbHksIG9ubHkgYSB2ZXJ5IHNtYWxsDQogICBzdWJzZXQgb2YgdGhlIHJvdXRlcnMgaW4g dGhlIG5ldHdvcmsgYXJlIHBhcnRpY2lwYXRpbmcgaW4gYW55IGdpdmVuDQogICBtdWx0aWNhc3Qg Z3JvdXAuICBJbiB0aGVzZSBjYXNlcywgdGhlIGJpdHN0cmluZyBlbmNvZGluZyB1c2VkIGluIEJJ RVINCiAgIHdvdWxkIGVuZCB1cCBoYXZpbmcgdmVyeSBmZXcgb25lcyBpbiBiZXR3ZWVuIG1hbnkg emVyb3MsIHJlc3VsdGluZyBpbg0KICAgaW5lZmZpY2llbnQgZW5jb2RpbmcuICBJbiBjYXNlIHdo ZW4gZ3JvdXAgbWVtYmVyc2hpcCBzcGFuIGFjcm9zcw0KICAgbXVsdGlwbGUgcmVnaW9ucyAoYW5k IGNhcnJpZXMgZGlmZmVyZW50IFNJcyksIGluZ3Jlc3MgcmVwbGljYXRpb24NCiAgIHdpbGwgYmUg dXNlZCwgY3JlYXRpbmcgZXZlbiBtb3JlIGluZWZmaWNhY3kgZHVlIHRvIHBhY2tldA0KICAgcmVw bGljYXRpb25zIGF0IGluZ3Jlc3MuDQoNCg0KDQpUaWFuLCBldCBhbC4gICAgICAgICAgICAgICBF eHBpcmVzIE1heSA5LCAyMDE1ICAgICAgICAgICAgICAgICAgW1BhZ2UgMl0NCgwNCkludGVybmV0 LURyYWZ0ICAgICAgICBCbG9jayBCYXNlZCBFbmNvZGluZyBmb3IgQklFUiAgICAgICAgTm92ZW1i ZXIgMjAxNA0KDQoNCjEuMS4gIFJlcXVpcmVtZW50cyBMYW5ndWFnZQ0KDQogICBUaGUga2V5IHdv cmRzICJNVVNUIiwgIk1VU1QgTk9UIiwgIlJFUVVJUkVEIiwgIlNIQUxMIiwgIlNIQUxMIE5PVCIs DQogICAiU0hPVUxEIiwgIlNIT1VMRCBOT1QiLCAiUkVDT01NRU5ERUQiLCAiTUFZIiwgYW5kICJP UFRJT05BTCIgaW4gdGhpcw0KICAgZG9jdW1lbnQgYXJlIHRvIGJlIGludGVycHJldGVkIGFzIGRl c2NyaWJlZCBpbiBSRkMgMjExOSBbUkZDMjExOV0uDQoNCjEuMi4gIEFjcm9ueW1zDQoNCiAgIEJJ RVIgICAtICBCaXQgSW5kZXggRXhwbGljaXQgUmVwbGljYXRpb24NCg0KICAgQkZSICAgIC0gIEJp dCBGb3J3YXJkaW5nIFJvdXRlcg0KDQoyLiAgUHJvcG9zZWQgU29sdXRpb24NCg0KICAgVGhlIHBy b3Bvc2VkIHNvbHV0aW9uIGlzIHRvIGxldmVyYWdlIHRoZSBmYWN0IHRoYXQgbWFueSBwYXJ0cyBv ZiB0aGUNCiAgIGJpdHN0cmluZyBhcmUgemVyb3MuICBJbiB0aGlzIHNvbHV0aW9uIHdlIGFsc28g ZGl2aWRlIHRoZSBuZXR3b3JrDQogICBpbnRvIHJlZ2lvbnMgYW5kIHVzZSBTSSB0byBpZGVudGlm eSB0aGUgcmVnaW9uLiAgSW4gb3VyIGVuY29kaW5nLCB3ZQ0KICAgd2lzaCB0byBpbmNsdWRlIG9u bHkgdGhlIHJlZ2lvbnMgdGhhdCBjb250YWluIGdyb3VwIG1lbWJlcnM7IG9yIGluDQogICBvdGhl ciB3b3JkcywgdG8gaW5jbHVkZSBvbmx5IHRoZSByZWdpb25zIHdob3NlIGNvcnJlc3BvbmRpbmcN CiAgIGJpdHN0cmluZyBhcmUgbm9uLXplcm8uICBXZSBsZWF2ZSBvdXQgdGhlIHJlZ2lvbnMgKGFu ZCB0aGUNCiAgIGNvcnJlc3BvbmRpbmcgU3ViQml0U3RyaW5ncykgdGhhdCBoYXZlIG5vIGdyb3Vw IG1lbWJlcnMuDQoNCjIuMS4gIE5ldyBFbmNvZGluZw0KDQogICBUaGUgZnVsbCBCRlItaWQgaXMg Y29tcG9zZWQgb2YgaGlnaGVyIG9yZGVyIFMtYml0cyBvZiBTSSBhbmQgbG93ZXINCiAgIG9yZGVy IEItYml0cyBvZiBTdWJCRlItaWQuICBUaGUgcGFyYW1ldGVycyBTIGFuZCBCIGNhbiBiZSBhc3Np Z25lZA0KICAgZGlmZmVyZW50IHZhbHVlcyBkZXBlbmRpbmcgb24gdGhlIGNvbmZpZ3VyYXRpb24u DQoNCiAgICstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rDQogICB8IEhpZ2ggT3JkZXIgU0kgKFMgYml0cykgICAgICAgIHwgTG93 IE9yZGVyIFN1YkJGUi1pZCAoQiBiaXRzKSAgfA0KICAgKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSsNCg0KICAgICAgICAgICAg ICAgICAgICAgICAgRmlndXJlIDE6IEJGUi1pZCBzdHJ1Y3R1cmUNCg0KICAgQSBiaXRTdHJpbmcg b2YgbGVuZ3RoIFggYml0cyBpbiB0aGUgcGFja2V0IGhlYWRlciBjYW4gYmUgdXNlZCB0bw0KICAg ZW5jb2RlIHRoZSBzZXQgb2YgcmVjZWl2ZXJzIG9mIHRoZSBncm91cC4gIEhlcmUgdGhlIGJpdFN0 cmluZyBjYW4gYmUNCiAgIGRpdmlkZWQgaW50byBLIGJsb2Nrcywgb25lIGZvciBlYWNoIHJlZ2lv biB0aGF0IGhhcyByZWNlaXZlcnMgaW4gdGhlDQogICBncm91cC4gIEVhY2ggYmxvY2sgc3RhcnRz IHdpdGggUy1iaXQgU0ksIGZvbGxvd2VkIGJ5IDJeQiAoMiB0byB0aGUNCiAgIHBvd2VyIG9mIEIp IGJpdHMgU3ViQml0U3RyaW5nLg0KDQoyLjEuMS4gIEV4YW1wbGUgT25lDQoNCiAgIEluIHRoaXMg ZXhhbXBsZSwgd2Ugd2lsbCBkaXZpZGUgYSBuZXR3b3JrIG9mIG1heGltdW0gNDA5NiBCRlJzIGlu dG8NCiAgIDI1NiByZWdpb25zLCBlYWNoIGNvbnRhaW5zIDE2IEJGUnMuICBJbiB0aGlzIGNhc2Us IFM9OCwgQj00LCB3aGlsZQ0KICAgdGhlIGZ1bGwgQkZSLWlkIChTSStTdWJCRlItaWQpIGhhcyA4 KzQ9MTIgYml0cyBpbiB1c2UuICBJbiB0aGlzDQogICBleGFtcGxlLCB3ZSB1c2UgYSB0b3RhbCBv ZiAyNDAtYml0IGJpdFN0cmluZyAoWD0yNDApIGluIGVhY2ggcGFja2V0Lg0KICAgVGhlIGJpdFN0 cmluZyBpcyBkaXZpZGVkIGludG8gMTAgYmxvY2tzLiAgRWFjaCBibG9jayBpcyAyNC1iaXRzIGxv bmcsDQogICB3aGljaCBzdGFydHMgd2l0aCA4LWJpdCBTSSBmb2xsb3dlZCBieSAyXjQ9MTYgYml0 IFN1YkJpdFN0cmluZywgYXMNCg0KDQoNClRpYW4sIGV0IGFsLiAgICAgICAgICAgICAgIEV4cGly ZXMgTWF5IDksIDIwMTUgICAgICAgICAgICAgICAgICBbUGFnZSAzXQ0KDA0KSW50ZXJuZXQtRHJh ZnQgICAgICAgIEJsb2NrIEJhc2VkIEVuY29kaW5nIGZvciBCSUVSICAgICAgICBOb3ZlbWJlciAy MDE0DQoNCg0KICAgc2hvd24gYmVsb3cuICBUaGUgZ3JvdXAgbWVtYmVycyB3aXRoaW4gdGhlIHJl Z2lvbiBhcmUgZW5jb2RlZCBpbiB0aGUNCiAgIGNvcnJlc3BvbmRpbmcgU3ViQml0U3RyaW5nLiAg UGxlYXNlIG5vdGUgdGhhdCB0aGUgYmxvY2tzIHN0YXJ0IGFmdGVyDQogICB0aGUgQklFUiBoZWFk ZXIgaW4gdGhpcyBleGFtcGxlLg0KDQogICAgMCAgICAgICAgICAgICAgICAgICAxICAgICAgICAg ICAgICAgICAgIDIgICAgICAgICAgICAgICAgICAgMw0KICAgIDAgMSAyIDMgNCA1IDYgNyA4IDkg MCAxIDIgMyA0IDUgNiA3IDggOSAwIDEgMiAzIDQgNSA2IDcgOCA5IDAgMQ0KICAgKy0rLSstKy0r LSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSsN CiAgIHwwIDAgMCAwfCAgVmVyICB8SXwwIDAgMCAwIDAgMCAwfCBQcm90byB8ICBMZW4gIHwgICAg RW50cm9weSAgICB8DQogICArLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKy0rLSstKw0KICAgfCBTSSAxICg4IGJpdHMpIHwgU3ViQml0U3Ry aW5nICAxICggMTYgYml0cykgICAgfA0KICAgKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKw0KICAgfiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgfg0KICAgKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKw0KICAgfCBTSSAxMCAoOCBiaXRzKXwgU3ViQml0U3RyaW5nIDEwICggMTYg Yml0cykgICAgfA0KICAgKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKw0KDQogICAgICAgICAgICAgICAgRmlndXJlIDI6IEVuY29kaW5nIEV4YW1wbGUgMSAo MjQtYml0IGJsb2NrKQ0KDQogICBJbiB0aGlzIGNhc2UsIHRoZSBncm91cCBjYW4gaGF2ZSBtZW1i ZXJzIGZyb20gMTAgZGlmZmVyZW50IHJlZ2lvbnMsDQogICBlYWNoIHJlZ2lvbiBjYW4gaGF2ZSB1 cCB0byAxNiBtZW1iZXJzLiAgVGhlcmVmb3JlIHRoZSBtYXggZ3JvdXANCiAgIG1lbWJlcnNoaXAg Y2FuIGJlIGVuY29kZWQgaW4gdGhlIDI0MC1iaXQgYml0U3RyaW5nIGlzIDEwKjE2PTE2MCBCRlJz LA0KICAgb3V0IG9mIHRoZSB0b3RhbCA0MDk2LiAgTW9yZSBncm91cCBtZW1iZXJzIGNhbiBiZSBh Y2NvbW9kYXRlZCBieQ0KICAgYWRkaW5nIG1vcmUgYmxvY2tzLg0KDQoyLjEuMi4gIEV4YW1wbGUg VHdvDQoNCiAgIE90aGVyIHZhcmlhdGlvbnMgYXJlIGFsc28gcG9zc2libGUuICBGb3IgZXhhbXBs ZSwgZm9yIGEgZG9tYWluIG9mDQogICBtYXhpbXVtIDgxOTIgQkZSUywgd2UgY2FuIGhhdmUgUz04 IGFuZCBCPTUsIGNvcnJlc3BvbmRpbmcgdG8gMjU2IHNldHMNCiAgIG9mIDMyIEJGUnMgZWFjaC4g IFRoZSBmdWxsIEJGUi1pZCAoU0krU3ViQkZSLWlkKSBoZXJlIGhhcyA4KzU9MTMgYml0cw0KICAg aW4gdXNlLiAgQWdhaW4gd2UgY2FuIHVzZSBhIDI0MC1iaXQgYml0U3RyaW5nIChYPTI0MCksIHdo aWNoIGlzDQogICBkaXZpZGVkIGludG8gNiBibG9ja3Mgb2YgNDAgYml0cyBlYWNoLiAgRWFjaCBi bG9jayBzdGFydHMgd2l0aCA4LWJpdA0KICAgU0ksIGZvbGxvd2VkIGJ5IDJeNT0zMiBiaXRzIG9m IFN1YkJpdFN0cmluZy4gIFBsZWFzZSBub3RlIHRoYXQgdGhlDQogICBibG9ja3MgaXMgc2hvd24g aW1tZWRpYXRlbHkgYWZ0ZXIgdGhlIEJJRVIgaGVhZGVyIGluIHRoaXMgZXhhbXBsZS4NCg0KDQoN Cg0KDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0KDQoNClRpYW4sIGV0IGFsLiAgICAgICAgICAgICAg IEV4cGlyZXMgTWF5IDksIDIwMTUgICAgICAgICAgICAgICAgICBbUGFnZSA0XQ0KDA0KSW50ZXJu ZXQtRHJhZnQgICAgICAgIEJsb2NrIEJhc2VkIEVuY29kaW5nIGZvciBCSUVSICAgICAgICBOb3Zl bWJlciAyMDE0DQoNCg0KICAgIDAgICAgICAgICAgICAgICAgICAgMSAgICAgICAgICAgICAgICAg ICAyICAgICAgICAgICAgICAgICAgIDMNCiAgICAwIDEgMiAzIDQgNSA2IDcgOCA5IDAgMSAyIDMg NCA1IDYgNyA4IDkgMCAxIDIgMyA0IDUgNiA3IDggOSAwIDENCiAgICstKy0rLSstKy0rLSstKy0r LSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rDQogICB8MCAw IDAgMHwgIFZlciAgfEl8MCAwIDAgMCAwIDAgMHwgUHJvdG8gfCAgTGVuICB8ICAgIEVudHJvcHkg ICAgfA0KICAgKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSsNCiAgIHwgU0kgMSAoOCBiaXRzKSB8DQogICArLSstKy0rLSstKy0r LSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKw0KICAg fCBTdWJCaXRTdHJpbmcgIDEgKCAzMiBiaXRzKSAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgIHwNCiAgICstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKy0rDQogICB+ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgfg0KICAgKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSsNCiAgIHwgU0kg NiAoOCBiaXRzKSB8DQogICArLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKy0rLSstKw0KICAgfCBTdWJCaXRTdHJpbmcgIDYgKCAzMiBiaXRz KSAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgIHwNCiAgICstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rDQoNCiAg ICAgICAgICAgICAgICBGaWd1cmUgMzogRW5jb2RpbmcgRXhhbXBsZSAyICg0MC1iaXQgYmxvY2sp DQoNCiAgIEluIHRoaXMgY2FzZSwgdGhlIGdyb3VwIGNhbiBoYXZlIG1lbWJlcnMgZnJvbSA2IGRp ZmZlcmVudCByZWdpb25zLA0KICAgZWFjaCByZWdpb24gY2FuIGhhdmUgdXAgdG8gMzIgbWVtYmVy cy4gIFRoZXJlZm9yZSB0aGUgbWF4IGdyb3VwDQogICBtZW1iZXJzaGlwIHRoYXQgY2FuIGJlIGVu Y29kZWQgaW4gdGhlIDI0MC1iaXQgYml0U3RyaW5nIGlzIDYqMzI9MTkyDQogICBCRlJzLCBvdXQg b2YgdG90YWwgb2YgODE5Mi4gIEFnYWluIG1vcmUgZ3JvdXAgbWVtYmVycyBjYW4gYmUNCiAgIGFj Y29tbW9kYXRlZCBieSBhZGRpbmcgbW9yZSBibG9ja3MuDQoNCjIuMi4gIFBhY2tldCBQcm9jZXNz aW5nIEZsb3cNCg0KICAgVGhlIHBhY2tldCBmb3J3YXJkaW5nIHByb2Nlc3MgZm9yIHRoaXMgZW5j b2RpbmcgaXMgc2ltaWxhciB0byB0aGUNCiAgIGNvbmNlcHQgZGVzY3JpYmVkIGluIFtJLUQuZHJh ZnQtd2lqbmFuZHMtYmllci1hcmNoaXRlY3R1cmVdLg0KDQogICBGb2xsb3dpbmcgaXMgdGhlIHBz ZXVkbyBjb2RlIGRlc2NyaWJpbmcgdGhlIHBhY2tldCBwcm9jZXNzaW5nDQogICBjb25lY3B0cyB1 c2luZyB0aGUgc2V0dXAgYXMgZGVzY3JpYmVkIGluIGV4YW1wbGUgb25lIChTZWN0aW9uIDIuMS4x KS4NCiAgIEhlcmUgZWFjaCBibG9jayBpbiB0aGUgZW5jb2RpbmcgaXMgMjQtYml0cywgc3RhcnRp bmcgd2l0aCA4LWJpdCBTSSwNCiAgIHRoZW4gZm9sbG93ZWQgYnkgMTYtYml0IFN1YkJpdFN0cmlu Zy4NCg0KICAgQml0TWFza0FuZCgpIGFuZCBCaXRNYXNrQ2xlYXIoKSBmdW5jdGlvbnMgcGVyZm9y bSBiaXQtd2lzZSBvcGVyYXRpb25zDQogICBvbiB0aGUgU3ViQml0U3RyaW5ncyBpbiB0aGUgZW5j b2RlZCBibG9ja3MsIHdpdGggY29ycmVzcG9uZGluZw0KICAgU3ViQml0U3RyaW5ncyBmcm9tIEZ1 bGxCaXRNYXNrLg0KDQogICAoU2VlIG5leHQgcGFnZS4pDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0K DQpUaWFuLCBldCBhbC4gICAgICAgICAgICAgICBFeHBpcmVzIE1heSA5LCAyMDE1ICAgICAgICAg ICAgICAgICAgW1BhZ2UgNV0NCgwNCkludGVybmV0LURyYWZ0ICAgICAgICBCbG9jayBCYXNlZCBF bmNvZGluZyBmb3IgQklFUiAgICAgICAgTm92ZW1iZXIgMjAxNA0KDQoNCiNkZWZpbmUgTUFYX0JM T0NLUyAxMA0Kc3RydWN0IFN1YkJpdFN0cmluZ19zIHsNCiAgICB1X2ludDhfdCB1Y2hhcnNbMl07 IC8vIDE2IGJpdHMNCn07DQpTdWJCaXRTdHJpbmdMZW5ndGggPSAxNjsNCg0Kc3RydWN0IEJsb2Nr X3Mgew0KICAgIHVfaW50OF90IFNJOyAgLy8gOCBiaXRzDQogICAgc3RydWN0IFN1YkJpdFN0cmlu Z19zIFN1YkJpdFN0cmluZzsgLy8gMTYgYml0cw0KfTsNCnN0cnVjdCBQYWNrZXRfcyB7DQogICAg c3RydWN0IEJsb2NrX3MgQmxvY2tzW01BWF9CTE9DS1NdOw0KICAgIC8vIFBheUxvYWQgc3RhcnQg aGVyZQ0KfTsNCnN0cnVjdCBGdWxsQml0TWFza19zIHsNCiAgICBzdHJ1Y3QgU3ViQml0U3RyaW5n X3MgU3ViQml0U3RyaW5nc1syNTZdOyAvLyAxNioyNTY9NDA5NiBiaXRzDQp9Ow0Kdm9pZCBCaXRN YXNrQW5kKFBhY2tldCwgRnVsbEJpdE1hc2spDQp7DQogICAgZm9yIChpPTA7IGk8TUFYX0JMT0NL UzsgaSsrKSB7DQogICAgICAgIFNJPVBhY2tldC0+QmxvY2tzW2ldLlNJOw0KICAgICAgICBQYWNr ZXQtPkJsb2Nrc1tpXS5TdWJCaXRTdHJpbmcgJj0gRnVsbEJpdE1hc2stPlN1YkJpdFN0cmluZ3Nb U0ldOw0KICAgIH0NCn0NCnZvaWQgQml0TWFza0NsZWFyKFBhY2tldCwgRnVsbEJpdE1hc2spDQp7 DQogICAgZm9yIChpPTA7IGk8TUFYX0JMT0NLUzsgaSsrKSB7DQogICAgICAgIFNJPVBhY2tldC0+ QmxvY2tzW2ldLlNJOw0KICAgICAgICBQYWNrZXQtPkJsb2Nrc1tpXS5TdWJCaXRTdHJpbmcgJj0g fkZ1bGxCaXRNYXNrLT5TdWJCaXRTdHJpbmdzW1NJXTsNCiAgICB9DQp9DQp2b2lkIEZvcndhcmRC aXRNYXNrUGFja2V0IChQYWNrZXQpDQp7DQogICAgZm9yIChpPTA7IGk8TUFYX0JMT0NLUzsgKysp IHsNCiAgICAgICAgU0k9UGFja2V0LT5CbG9ja3NbaV0uU0k7DQogICAgICAgIE9mZnNldD1TSSpT dWJCaXRTdHJpbmdMZW5ndGg7DQogICAgICAgIFN1YkJpdFN0cmluZyA9IFBhY2tldC0+QmxvY2tz W2ldLlN1YkJpdFN0cmluZzsNCiAgICAgICAgZm9yIChJbmRleCA9IEdldEZpcnN0Qml0UG9zaXRp b24oU3ViQml0U3RyaW5nKTsgSW5kZXggOw0KICAgICAgICAgICAgIEluZGV4ID0gR2V0TmV4dEJp dFBvc2l0aW9uKFN1YkJpdFN0cmluZywgSW5kZXgpKSB7DQogICAgICAgIHN0cnVjdCBGdWxsQml0 TWFza19zICpGLUJNID0gQklGVFtJbmRleCtPZmZzZXRdLT5GLUJNOw0KICAgICAgICBpZiAoIUYt Qk0pIGNvbnRpbnVlOw0KICAgICAgICBCRlItTkJSID0gQklGVFtJbmRleCtPZmZzZXRdLT5CRlIt TkJSOw0KICAgICAgICBQYWNrZXRDb3B5ID0gQ29weShQYWNrZXQpOw0KICAgICAgICBCaXRNYXNr QW5kKFBhY2tldENvcHksIEYtQk0pOw0KICAgICAgICBQYWNrZXRTZW5kKFBhY2tldENvcHksIEJG Ui1OQlIpOw0KICAgICAgICBCaXRNYXNrQ2xlYXIoUGFja2V0LCBGLUJNKTsNCiAgICB9DQp9DQoN Cg0KDQpUaWFuLCBldCBhbC4gICAgICAgICAgICAgICBFeHBpcmVzIE1heSA5LCAyMDE1ICAgICAg ICAgICAgICAgICAgW1BhZ2UgNl0NCgwNCkludGVybmV0LURyYWZ0ICAgICAgICBCbG9jayBCYXNl ZCBFbmNvZGluZyBmb3IgQklFUiAgICAgICAgTm92ZW1iZXIgMjAxNA0KDQoNCiAgICAgICAgICAg ICAgICAgIEZpZ3VyZSA0OiBQYWNrZXQgUHJvY2Vzc2luZyBQc2V1ZG8gQ29kZQ0KDQozLiAgQ29u Y2x1c2lvbnMNCg0KICAgSW4gdGhpcyBkb2N1bWVudCB3ZSBwcm9wb3NlIGEgbmV3IHBhY2tldCBl bmNvZGluZyBzY2hlbWUgZm9yIEJJRVINCiAgIHRoYXQgY2FuIGltcHJvdmUgdGhlIGVuY29kaW5n IGVmZmljaWVuY3kgaW4gbGFyZ2UgbmV0d29ya3MsIHdoZW4NCiAgIGdyb3VwIG1lbWJlcnMgYXJl IHNwYXJzZWx5IHBvcHVsYXRlZC4gIFRoZSBuZXcgc2NoZW1lIGFsc28gY2FuIGF2b2lkDQogICB1 bm5lY2Vzc2FyeSBwYWNrZXQgcmVwbGljYXRpb24gYXQgaW5ncmVzcyB3aGVuIG1lbWJlcnMgb2Yg YSBzcGFyc2VseQ0KICAgcG9wdWxhdGVkIG11bHRpY2FzdCBncm91cCBzcGFuIGFjcm9zcyBtdWx0 aXBsZSBkaWZmZXJlbnQgcmVnaW9pbnMNCiAgIChoYXZpbmcgZGlmZmVyZW50IFNJcykuDQoNCjQu ICBJQU5BIENvbnNpZGVyYXRpb25zDQoNCiAgIFRoaXMgZG9jdW1lbnQgZGVmaW5lcyBubyBuZXcg bmFtZXNwYWNlcy4NCg0KNS4gIFNlY3VyaXR5IENvbnNpZGVyYXRpb25zDQoNCiAgIFRoaXMgcHJv cG9zYWwgYnJvdWdodCBubyBuZXcgc2VjdXJpdHkgY29uY2VybnMgb3RoZXIgdGhhbiB0aG9zZQ0K ICAgaWRlbnRpZmllZCBpbiBbSS1ELmRyYWZ0LXdpam5hbmRzLWJpZXItYXJjaGl0ZWN0dXJlXS4N Cg0KNi4gIFJlZmVyZW5jZXMNCg0KNi4xLiAgTm9ybWF0aXZlIFJlZmVyZW5jZXMNCg0KICAgW1JG QzIxMTldICBCcmFkbmVyLCBTLiwgIktleSB3b3JkcyBmb3IgdXNlIGluIFJGQ3MgdG8gSW5kaWNh dGUNCiAgICAgICAgICAgICAgUmVxdWlyZW1lbnQgTGV2ZWxzIiwgQkNQIDE0LCBSRkMgMjExOSwg TWFyY2ggMTk5Ny4NCg0KNi4yLiAgSW5mb3JtYXRpdmUgUmVmZXJlbmNlcw0KDQogICBbSS1ELndp am5hbmRzLWJpZXItYXJjaGl0ZWN0dXJlXQ0KICAgICAgICAgICAgICBXaWpuYW5kcywgSS4sIFJv c2VuLCBFLiwgRG9sZ2Fub3csIEEuLCBQcnp5Z2llbmRhLCBULiwgYW5kDQogICAgICAgICAgICAg IFMuIEFsZHJpbiwgIk11bHRpY2FzdCB1c2luZyBCaXQgSW5kZXggRXhwbGljaXQNCiAgICAgICAg ICAgICAgUmVwbGljYXRpb24iLCBkcmFmdC13aWpuYW5kcy1iaWVyLWFyY2hpdGVjdHVyZS0wMSAo d29yayBpbg0KICAgICAgICAgICAgICBwcm9ncmVzcyksIE9jdG9iZXIgMjAxNC4NCg0KICAgW0kt RC53aWpuYW5kcy1tcGxzLWJpZXItZW5jYXBzdWxhdGlvbl0NCiAgICAgICAgICAgICAgV2lqbmFu ZHMsIEkuLCBSb3NlbiwgRS4sIERvbGdhbm93LCBBLiwgVGFudHN1cmEsIEouLCBhbmQNCiAgICAg ICAgICAgICAgUy4gQWxkcmluLCAiRW5jYXBzdWxhdGlvbiBmb3IgQml0IEluZGV4IEV4cGxpY2l0 DQogICAgICAgICAgICAgIFJlcGxpY2F0aW9uIGluIE1QTFMgTmV0d29ya3MiLCBkcmFmdC13aWpu YW5kcy1tcGxzLWJpZXItDQogICAgICAgICAgICAgIGVuY2Fwc3VsYXRpb24tMDEgKHdvcmsgaW4g cHJvZ3Jlc3MpLCBPY3RvYmVyIDIwMTQuDQoNCkF1dGhvcnMnIEFkZHJlc3Nlcw0KDQoNCg0KDQoN Cg0KDQoNCg0KVGlhbiwgZXQgYWwuICAgICAgICAgICAgICAgRXhwaXJlcyBNYXkgOSwgMjAxNSAg ICAgICAgICAgICAgICAgIFtQYWdlIDddDQoMDQpJbnRlcm5ldC1EcmFmdCAgICAgICAgQmxvY2sg QmFzZWQgRW5jb2RpbmcgZm9yIEJJRVIgICAgICAgIE5vdmVtYmVyIDIwMTQNCg0KDQogICBBbGJl cnQgVGlhbg0KICAgRXJpY3Nzb24sIEluYy4NCiAgIDI1MCBIb2xnZXIgV2F5LA0KICAgU2FuIEpv c2UsIENhbGlmb3JuaWEgIDk1MTM0DQogICBVU0ENCg0KICAgRW1haWw6IGFsYmVydC50aWFuQGVy aWNzc29uLmNvbQ0KDQoNCiAgIFJhbWFuYXRoYW4gTGFrc2htaWthbnRoYW4NCiAgIEVyaWNzc29u LCBJbmMuDQogICAyNTAgSG9sZ2VyIFdheSwNCiAgIFNhbiBKb3NlLCBDYWxpZm9ybmlhICA5NTEz NA0KICAgVVNBDQoNCiAgIEVtYWlsOiByYW1hbmF0aGFuLmxha3NobWlrYW50aGFuQGVyaWNzc29u LmNvbQ0KDQoNCiAgIEplZmYgVGFudHN1cmENCiAgIEVyaWNzc29uLCBJbmMuDQogICAyMDAgSG9s Z2VyIFdheSwNCiAgIFNhbiBKb3NlLCBDYWxpZm9ybmlhICA5NTEzNA0KICAgVVNBDQoNCiAgIEVt YWlsOiBqZWZmLnRhbnRzdXJhQGVyaWNzc29uLmNvbQ0KDQoNCiAgIEFudG9uaSBQcnp5Z2llbmRh DQogICBFcmljc3NvbiwgSW5jLg0KICAgMzAwIEhvbGdlciBXYXksDQogICBTYW4gSm9zZSwgQ2Fs aWZvcm5pYSAgOTUxMzQNCiAgIFVTQQ0KDQogICBFbWFpbDogYW50b25pLnByenlnaWVuZGFAZXJp Y3Nzb24uY29tDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0KDQoNCg0KVGlhbiwgZXQgYWwu ICAgICAgICAgICAgICAgRXhwaXJlcyBNYXkgOSwgMjAxNSAgICAgICAgICAgICAgICAgIFtQYWdl IDhdDQo= --_005_D0812D8D311743AlbertTianericssoncom_ Content-Type: text/plain; name="draft-tian-bier-list-00.txt" Content-Description: draft-tian-bier-list-00.txt Content-Disposition: attachment; filename="draft-tian-bier-list-00.txt"; size=11176; creation-date="Thu, 06 Nov 2014 21:52:14 GMT"; modification-date="Thu, 06 Nov 2014 21:52:14 GMT" Content-ID: <7B806F6B1962B240BEC730DCEDA9226F@ericsson.com> Content-Transfer-Encoding: base64 CgoKCldvcmtpbmcgR3JvdXAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgQS4gVGlhbgpJbnRlcm5ldC1EcmFmdCAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgIFIuIExha3NobWkKSW50ZW5kZWQgc3RhdHVzOiBTdGFu ZGFyZHMgVHJhY2sgICAgICAgICAgICAgICAgICAgICAgICAgICAgIEouIFRhbnRzdXJhCkV4cGly ZXM6IE1heSA5LCAyMDE1ICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgQS4g UHJ6eWdpZW5kYQogICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgRXJpY3Nzb24sIEluYy4KICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgICAgICAgICAgICAgICBOb3ZlbWJlciA1LCAyMDE0CgoKICAgICAgICAgICAg ICAgICAgICAgIExpc3QgQmFzZWQgRW5jb2RpbmcgZm9yIEJJRVIKICAgICAgICAgICAgICAgICAg ICAgICAgZHJhZnQtdGlhbi1iaWVyLWxpc3QtMDAKCkFic3RyYWN0CgogICBUaGlzIGRvY3VtZW50 IGRlc2NyaWJlcyBhIG5ldyBleHBsaWNpdCBsaXN0IGJhc2VkIGVuY29kaW5nIHNjaGVtZSBmb3IK ICAgQml0IEluZGV4IEV4cGxpY2l0IFJlcGxpY2F0aW9uIChCSUVSKSBtdWx0aWNhc3QgZm9yd2Fy ZGluZyBzb2x1dGlvbgogICB0byBiZXR0ZXIgaGFuZGxlIHNwYXJzZWx5IHBvcHVsYXRlZCBtdWx0 aWNhc3QgZ3JvdXBzIGluIGxhcmdlCiAgIG5ldHdvcmtzLiAgVGhlIGFzc29jaWF0ZWQgcGFja2V0 IHByb2Nlc3NpbmcgaXMgYWxzbyBkZXNjcmliZWQuCgpTdGF0dXMgb2YgVGhpcyBNZW1vCgogICBU aGlzIEludGVybmV0LURyYWZ0IGlzIHN1Ym1pdHRlZCBpbiBmdWxsIGNvbmZvcm1hbmNlIHdpdGgg dGhlCiAgIHByb3Zpc2lvbnMgb2YgQkNQIDc4IGFuZCBCQ1AgNzkuCgogICBJbnRlcm5ldC1EcmFm dHMgYXJlIHdvcmtpbmcgZG9jdW1lbnRzIG9mIHRoZSBJbnRlcm5ldCBFbmdpbmVlcmluZwogICBU YXNrIEZvcmNlIChJRVRGKS4gIE5vdGUgdGhhdCBvdGhlciBncm91cHMgbWF5IGFsc28gZGlzdHJp YnV0ZQogICB3b3JraW5nIGRvY3VtZW50cyBhcyBJbnRlcm5ldC1EcmFmdHMuICBUaGUgbGlzdCBv ZiBjdXJyZW50IEludGVybmV0LQogICBEcmFmdHMgaXMgYXQgaHR0cDovL2RhdGF0cmFja2VyLmll dGYub3JnL2RyYWZ0cy9jdXJyZW50Ly4KCiAgIEludGVybmV0LURyYWZ0cyBhcmUgZHJhZnQgZG9j dW1lbnRzIHZhbGlkIGZvciBhIG1heGltdW0gb2Ygc2l4IG1vbnRocwogICBhbmQgbWF5IGJlIHVw ZGF0ZWQsIHJlcGxhY2VkLCBvciBvYnNvbGV0ZWQgYnkgb3RoZXIgZG9jdW1lbnRzIGF0IGFueQog ICB0aW1lLiAgSXQgaXMgaW5hcHByb3ByaWF0ZSB0byB1c2UgSW50ZXJuZXQtRHJhZnRzIGFzIHJl ZmVyZW5jZQogICBtYXRlcmlhbCBvciB0byBjaXRlIHRoZW0gb3RoZXIgdGhhbiBhcyAid29yayBp biBwcm9ncmVzcy4iCgogICBUaGlzIEludGVybmV0LURyYWZ0IHdpbGwgZXhwaXJlIG9uIE1heSA5 LCAyMDE1LgoKQ29weXJpZ2h0IE5vdGljZQoKICAgQ29weXJpZ2h0IChjKSAyMDE0IElFVEYgVHJ1 c3QgYW5kIHRoZSBwZXJzb25zIGlkZW50aWZpZWQgYXMgdGhlCiAgIGRvY3VtZW50IGF1dGhvcnMu ICBBbGwgcmlnaHRzIHJlc2VydmVkLgoKICAgVGhpcyBkb2N1bWVudCBpcyBzdWJqZWN0IHRvIEJD UCA3OCBhbmQgdGhlIElFVEYgVHJ1c3QncyBMZWdhbAogICBQcm92aXNpb25zIFJlbGF0aW5nIHRv IElFVEYgRG9jdW1lbnRzCiAgIChodHRwOi8vdHJ1c3RlZS5pZXRmLm9yZy9saWNlbnNlLWluZm8p IGluIGVmZmVjdCBvbiB0aGUgZGF0ZSBvZgogICBwdWJsaWNhdGlvbiBvZiB0aGlzIGRvY3VtZW50 LiAgUGxlYXNlIHJldmlldyB0aGVzZSBkb2N1bWVudHMKICAgY2FyZWZ1bGx5LCBhcyB0aGV5IGRl c2NyaWJlIHlvdXIgcmlnaHRzIGFuZCByZXN0cmljdGlvbnMgd2l0aCByZXNwZWN0CiAgIHRvIHRo aXMgZG9jdW1lbnQuICBDb2RlIENvbXBvbmVudHMgZXh0cmFjdGVkIGZyb20gdGhpcyBkb2N1bWVu dCBtdXN0CiAgIGluY2x1ZGUgU2ltcGxpZmllZCBCU0QgTGljZW5zZSB0ZXh0IGFzIGRlc2NyaWJl ZCBpbiBTZWN0aW9uIDQuZSBvZgoKCgoKVGlhbiwgZXQgYWwuICAgICAgICAgICAgICAgRXhwaXJl cyBNYXkgOSwgMjAxNSAgICAgICAgICAgICAgICAgIFtQYWdlIDFdCgwKSW50ZXJuZXQtRHJhZnQg ICAgICAgIExpc3QgQmFzZWQgRW5jb2RpbmcgZm9yIEJJRVIgICAgICAgICBOb3ZlbWJlciAyMDE0 CgoKICAgdGhlIFRydXN0IExlZ2FsIFByb3Zpc2lvbnMgYW5kIGFyZSBwcm92aWRlZCB3aXRob3V0 IHdhcnJhbnR5IGFzCiAgIGRlc2NyaWJlZCBpbiB0aGUgU2ltcGxpZmllZCBCU0QgTGljZW5zZS4K ClRhYmxlIG9mIENvbnRlbnRzCgogICAxLiAgSW50cm9kdWN0aW9uICAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAgIDIKICAgICAxLjEuICBSZXF1aXJlbWVu dHMgTGFuZ3VhZ2UgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gICAzCiAgICAg MS4yLiAgQWNyb255bXMgIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gLiAuICAgMwogICAyLiAgUHJvcG9zZWQgU29sdXRpb24gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAgIDMKICAgICAyLjEuICBOZXcgRW5jb2RpbmcgIC4gLiAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gICAzCiAgICAgMi4yLiAgUGFj a2V0IFByb2Nlc3NpbmcgRmxvdyAgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuICAg NAogICAzLiAgQ29uY2x1c2lvbnMgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAgIDUKICAgNC4gIElBTkEgQ29uc2lkZXJhdGlvbnMgLiAuIC4gLiAuIC4g LiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gICA1CiAgIDUuICBTZWN1cml0eSBDb25zaWRl cmF0aW9ucyAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuICAgNQogICA2LiAg UmVmZXJlbmNlcyAgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAu IC4gLiAgIDUKICAgICA2LjEuICBOb3JtYXRpdmUgUmVmZXJlbmNlcyAgLiAuIC4gLiAuIC4gLiAu IC4gLiAuIC4gLiAuIC4gLiAuIC4gICA1CiAgICAgNi4yLiAgSW5mb3JtYXRpdmUgUmVmZXJlbmNl cyAgLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuICAgNQogICBBdXRob3JzJyBBZGRy ZXNzZXMgIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAuIC4gLiAgIDUK CjEuICBJbnRyb2R1Y3Rpb24KCiAgIEJpdCBJbmRleCBFeHBsaWNpdCBSZXBsaWNhdGlvbiAoQklF UiksIGFzIGRlZmluZWQgaW4gW0ktRC5kcmFmdC0KICAgd2lqbmFuZHMtYmllci1hcmNoaXRlY3R1 cmVdIGlzIGEgbmV3IGFyY2hpdGVjdHVyZSBmb3IgZm9yd2FyZGluZwogICBtdWx0aWNhc3QgZGF0 YSBwYWNrZXRzLCBieSBlbmNvZGluZyB0aGUgbXVsdGljYXN0IGdyb3VwIG1lbWJlcnNoaXAKICAg ZGlyZWN0bHkgaW4gYSBuZXcgcGFja2V0IGhlYWRlciB1c2luZyBhIGJpdFN0cmluZy4gIEJJRVIg YXNzaWduZXMgYW4KICAgdW5pcXVlIDE2LWJpdCBJRCAoQkZSLWlkKSB0byBlYWNoIEJpdCBGb3J3 YXJkaW5nIFJvdXRlciAoQkZSKSwKICAgcm91dGVycyB0aGF0IHN1cHBvcnRzIEJJRVIuICBCRlIt aWQgY29udGFpbnMgaGlnaGVyIDgtYml0IGFzIFNJCiAgICh3aGljaCBjYW4gYmUgdXNlZCB0byBy ZXByZXNlbnQgYSByZWdpb24pLCBhbmQgbG93ZXIgOC1iaXQgYXMgYSBiaXQKICAgcG9zaXRpb24g aW50byBhIDI1Ni1iaXQgYml0U3RyaW5nLiAgRWFjaCBzZXQgYml0IGluIHRoZSBiaXRTdHJpbmcK ICAgcmVwcmVzZW50IGEgbWVtYmVyIGluIHRoZSBncm91cCBmb3Igd2hpY2ggdGhlIHBhY2tldCBu ZWVkcyB0byBiZQogICBkZWxpdmVyZWQuICBCSUVSIHdvcmtzIHdlbGwgd2hlbiB0aGUgbmV0d29y ayBpcyBub3QgdG9vIGJpZyAobGVzcwogICB0aGFuIDI1NiBCRlJzKS4gIFdoZW4gdGhlIG5ldHdv cmsgaGFzIG1vcmUgdGhhbiAyNTYgQkZScywgbXVsdGlwbGUKICAgU0lzIHdpbGwgYmUgcmVxdWly ZWQuICBJdCBpcyBsaWtlbHkgdGhhdCB0aGUgbWVtYmVycyBvZiBhIGdyb3VwIGNhbgogICBzcGFu IGFjcm9zcyBtdWx0aXBsZSByZWdpb25zIGFuZCB0aGVyZWZvcmUgaGF2ZSBkaWZmZXJlbnQgU0lz LiAgSW4KICAgdGhpcyBjYXNlLCByZXBsaWNhdGlvbiBhdCB0aGUgaW5ncmVzcyB3aWxsIGJlIHVz ZWQsIG9uZSBwYWNrZXQgZm9yCiAgIGVhY2ggU0kgKHJlZ2lvbikuCgogICBJbiByZWFsaXR5LCBt b3N0IG9mIHRoZSBtdWx0aWNhc3QgZ3JvdXBzIGFyZSBzcGFyc2UsIHdoZW4gY29tcGFyZWQgdG8K ICAgdGhlIHRvdGFsIG51bWJlciBvZiByb3V0ZXJzIGluIGEgbmV0d29yay4gIFVzdWFsbHksIG9u bHkgYSB2ZXJ5IHNtYWxsCiAgIHN1YnNldCBvZiB0aGUgcm91dGVycyBpbiB0aGUgbmV0d29yayBh cmUgcGFydGljaXBhdGluZyBpbiBhbnkgZ2l2ZW4KICAgbXVsdGljYXN0IGdyb3VwLiAgSW4gdGhl c2UgY2FzZXMsIHRoZSBiaXRzdHJpbmcgZW5jb2RpbmcgdXNlZCBpbiBCSUVSCiAgIHdvdWxkIGVu ZCB1cCBoYXZpbmcgdmVyeSBmZXcgb25lcyBpbiBiZXR3ZWVuIG1hbnkgemVyb3MsIHJlc3VsdGlu ZyBpbgogICBpbmVmZmljaWVudCBlbmNvZGluZy4gIEluIGNhc2Ugd2hlbiBncm91cCBtZW1iZXJz aGlwIHNwYW4gYWNyb3NzCiAgIG11bHRpcGxlIHJlZ2lvbnMgKGFuZCBjYXJyaWVzIGRpZmZlcmVu dCBTSXMpLCBpbmdyZXNzIHJlcGxpY2F0aW9uCiAgIHdpbGwgYmUgdXNlZCwgY3JlYXRpbmcgZXZl biBtb3JlIGluZWZmaWNhY3kgZHVlIHRvIHBhY2tldAogICByZXBsaWNhdGlvbnMgYXQgaW5ncmVz cy4KCgoKCgpUaWFuLCBldCBhbC4gICAgICAgICAgICAgICBFeHBpcmVzIE1heSA5LCAyMDE1ICAg ICAgICAgICAgICAgICAgW1BhZ2UgMl0KDApJbnRlcm5ldC1EcmFmdCAgICAgICAgTGlzdCBCYXNl ZCBFbmNvZGluZyBmb3IgQklFUiAgICAgICAgIE5vdmVtYmVyIDIwMTQKCgoxLjEuICBSZXF1aXJl bWVudHMgTGFuZ3VhZ2UKCiAgIFRoZSBrZXkgd29yZHMgIk1VU1QiLCAiTVVTVCBOT1QiLCAiUkVR VUlSRUQiLCAiU0hBTEwiLCAiU0hBTEwgTk9UIiwKICAgIlNIT1VMRCIsICJTSE9VTEQgTk9UIiwg IlJFQ09NTUVOREVEIiwgIk1BWSIsIGFuZCAiT1BUSU9OQUwiIGluIHRoaXMKICAgZG9jdW1lbnQg YXJlIHRvIGJlIGludGVycHJldGVkIGFzIGRlc2NyaWJlZCBpbiBSRkMgMjExOSBbUkZDMjExOV0u CgoxLjIuICBBY3JvbnltcwoKICAgQklFUiAgIC0gIEJpdCBJbmRleCBFeHBsaWNpdCBSZXBsaWNh dGlvbgoKICAgQkZSICAgIC0gIEJpdCBGb3J3YXJkaW5nIFJvdXRlcgoKMi4gIFByb3Bvc2VkIFNv bHV0aW9uCgogICBJbiB0aGlzIGRvY3VtZW50IHdlIHByb3Bvc2UgYSBuZXcgZW5jb2Rpbmcgc2No ZW1lIHdoZXJlIHdlIGRpcmVjdGx5CiAgIGVuY29kZSBhIGxpc3Qgb2YgQkZSLWlkcyBpbiB0aGUg Yml0U3RyaW5nLiAgSW4gYSBzZXR1cCB3aXRoIEJpdFN0cmluZwogICBvZiBsZW5ndGggMjU2IGFu ZCBTSSByYW5nZSBvZiBbMC4uMjU1XSwgQkZSLWlkIHdvdWxkIGJlIDE2IGJpdHMgbG9uZwogICBh bmQgb25lIEJpdFN0cmluZyBjYW4gYWNjb21vZGF0ZSAxNiBzdWNoIEJGUi1pZHMuICBUaGUgbmV0 d29yayBjYW4KICAgaGF2ZSB1cCB0byA2NTUzNiBCRlJzLgoKICAgSW4gY2FzZSB0aGUgU0kgcmFu Z2UgaXMgWzAuLjE1XSwgZWFjaCBCRlItaWQgd291bGQgaGF2ZSBvbmx5IDEyIGJpdHMKICAgaW4g dXNlLCB0aGVyZWZvcmUgaWYgd2UgY2hvb3NlIHRvIHVzZSAxMi1iaXQgcGVyIEJGUi1pZCBlbmNv ZGluZywgMjEKICAgc3VjaCBzaG9ydGVuZWQgQkZSLWlkcyBjYW4gYmUgZW5jb2RlZCBpbiBhIEJp dFN0cmluZyBvZiBsZW5ndGggMjU2LgogICBUaGUgbmV0d29yayBjYW4gaGF2ZSB1cCB0byA0MDk2 IEJGUnMuCgoyLjEuICBOZXcgRW5jb2RpbmcKCiAgIEZvbGxvd2luZyBpcyBhbiBlbmNvZGluZyB1 c2luZyAxNi1iaXRzIHBlciBCRlItaWQgYXMgZXhhbXBsZS4KCiAgICAwICAgICAgICAgICAgICAg ICAgIDEgICAgICAgICAgICAgICAgICAgMiAgICAgICAgICAgICAgICAgICAzCiAgICAwIDEgMiAz IDQgNSA2IDcgOCA5IDAgMSAyIDMgNCA1IDYgNyA4IDkgMCAxIDIgMyA0IDUgNiA3IDggOSAwIDEK ICAgKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSsKICAgfDAgMCAwIDB8ICBWZXIgIHxJfDAgMCAwIDAgMCAwIDB8IFByb3RvIHwg IExlbiAgfCAgICBFbnRyb3B5ICAgIHwKICAgKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSsKICAgfCAgICBCRlItaWQgIDEgKGZp cnN0IDE2Yml0cykgICB8ICAgIEJGUi1pZCAgMiAobmV4dCAxNmJpdHMpICAgIHwKICAgKy0rLSst Ky0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0r LSsKICAgfiAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAgICAg ICAgICAgICAgICAgIH4KICAgKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0r LSstKy0rLSstKy0rLSstKy0rLSstKy0rLSsKICAgfCAgICBCRlItaWQgMTUgKFBlbnVsdGltYXRl KSAgICB8ICAgIEJGUi1pZCAxNiAobGFzdCAxNmJpdHMpICAgIHwKICAgKy0rLSstKy0rLSstKy0r LSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSstKy0rLSsKCiAgICAg ICAgICAgICAgICAgICAgICAgIEZpZ3VyZSAxOiBCRlItaWQgc3RydWN0dXJlCgogICBNb3JlIG9y IGxlc3MgZ3JvdXAgbWVtYmVycyBtYXkgYmUgYWNjb21tb2RhdGVkIGJ5IHNpbXBseSB1c2luZyBs b25nZXIKICAgb3Igc2hvcnRlciBiaXRTdHJpbmcsIGlmIHRoZSBvdmVyaGVhZCBpcyBkZWVtZWQg YWNjZXB0YWJsZS4KCgoKCgoKVGlhbiwgZXQgYWwuICAgICAgICAgICAgICAgRXhwaXJlcyBNYXkg OSwgMjAxNSAgICAgICAgICAgICAgICAgIFtQYWdlIDNdCgwKSW50ZXJuZXQtRHJhZnQgICAgICAg IExpc3QgQmFzZWQgRW5jb2RpbmcgZm9yIEJJRVIgICAgICAgICBOb3ZlbWJlciAyMDE0CgoKMi4y LiAgUGFja2V0IFByb2Nlc3NpbmcgRmxvdwoKICAgVGhlIHBhY2tldCBmb3J3YXJkaW5nIHByb2Nl c3MgZm9yIHRoaXMgZW5jb2RpbmcgaXMgc2ltaWxhciB0byB0aGUKICAgY29uY2VwdCBkZXNjcmli ZWQgaW4gW0ktRC5kcmFmdC13aWpuYW5kcy1iaWVyLWFyY2hpdGVjdHVyZV0uICBIZXJlCiAgIHRo ZSBTRlItaWRzIGFyZSByZWFkIGRpcmVjdGx5IGZyb20gdGhlIGJpdFN0cmluZywgd2l0aG91dCBi aXRTdHJpbmcKICAgdG8gYml0IHBvc2l0aW9uIGNvbnZlcnNpb24uCgogICBUaGUgcHJvY2VzcyBj YW4gYmUgcmVwcmVzZW50ZWQgYnkgdGhlIGZvbGxvd2luZyBwc2V1ZG8gY29kZS4gIEhlcmUgd2UK ICAgdXNlIDE2LWJpdHMgQkZSLWlkIGluIHRoaXMgZXhhbXBsZS4KCiNkZWZpbmUgTUFYX0xJU1Qg MTYKc3RydWN0IFBhY2tldF9zIHsKICAgIHVfaW50MTZfdCBCaXRTdHJpbmdBc0JGUmlkW01BWF9M SVNUXTsKICAgIC8vIFBheUxvYWQgc3RhcnQgaGVyZQp9OwovLyB0ZXN0IGlmIHRoZSBrJ3RoIGJp dCBpbiB0aGUgRi1CTSBpcyBzZXQsIGsgc3RhcnQgYXQgMQpib29sIEJpdE1hc2tUZXN0IChGLUJN LCBrKQp7CiAgICByZXR1cm4gKEYtQk0gJiAoMTw8KGstMSkpKTsKfQp2b2lkIEZvcndhcmRCaXRN YXNrUGFja2V0IChzdHJ1Y3QgUGFja2V0X3MgKlBhY2tldCkKewogICAgZm9yIChpID0gMDsgaSA8 IE1BWF9MSVNUOyBpKyspIHsKICAgICAgICBiZnJfaWRfaSA9IFBhY2tldC0+Qml0U3RyaW5nQXNC RlJpZFtpXTsKICAgICAgICBzdHJ1Y3QgRnVsbEJpdE1hc2tfcyAqRi1CTSA9IEJJRlRbYmZyX2lk X2ldLT5GLUJNOwogICAgICAgIGlmICghRi1CTSkgY29udGludWU7CiAgICAgICAgQkZSLU5CUiA9 IEJJRlRbYmZyX2lkXS0+QkZSLU5CUjsgLy8gZ2V0IE5CUiBpbmZvCiAgICAgICAgUGFja2V0Q29w eSA9IENvcHkoUGFja2V0KTsKICAgICAgICBQYWNrZXQtPkJpdFN0cmluZ0FzQkZSaWRbaV0gPSAw OyAvLyBiZnJfaWRfaSBpbiBjb3B5CiAgICAgICAgZm9yIChqPWkrMTsgajwgTUFYX0xJU1Q7IGor KykgewogICAgICAgICAgICBiZnJfaWRfaiA9IFBhY2tldC0+Qml0U3RyaW5nQXNCRlJpZFtqXTsK ICAgICAgICAgICAgaWYgKEJpdE1hc2tUZXN0KEYtQk0sIGJmcl9pZF9qKSkgeyAvLyBiaXQgZm9y IGJmcl9pZF9qIGlzIHNldAogICAgICAgICAgICAgICAgLy8gYmZyX2lkX2ogaW4gY29weSwgY2xl YXIgaW4gdGhlIG9yaWdpbmFsCiAgICAgICAgICAgICAgICBQYWNrZXQtPkJpdFN0cmluZ0FzQkZS aWRbal0gPSAwOwogICAgICAgICAgICB9IGVsc2UgewogICAgICAgICAgICAgICAgLy8gYmZyX2lk X2ogaW4gb3JpZ2luYWwsIGNsZWFyIGluIHRoZSBjb3B5CiAgICAgICAgICAgICAgICBQYWNrZXRD b3B5LT5CaXRTdHJpbmdBc0JGUmlkW2pdID0gMDsKICAgICAgICAgICAgfQogICAgICAgIH0KICAg ICAgICBQYWNrZXRTZW5kKFBhY2tldENvcHksIEJGUi1OQlIpOwogICAgfQp9CgogICAgICAgICAg ICAgICAgICAgICAgIFBhY2tldCBQcm9jZXNzaW5nIFBzZXVkbyBDb2RlCgoKCgoKClRpYW4sIGV0 IGFsLiAgICAgICAgICAgICAgIEV4cGlyZXMgTWF5IDksIDIwMTUgICAgICAgICAgICAgICAgICBb UGFnZSA0XQoMCkludGVybmV0LURyYWZ0ICAgICAgICBMaXN0IEJhc2VkIEVuY29kaW5nIGZvciBC SUVSICAgICAgICAgTm92ZW1iZXIgMjAxNAoKCjMuICBDb25jbHVzaW9ucwoKICAgSW4gdGhpcyBk b2N1bWVudCB3ZSBwcm9wb3NlIGEgbmV3IHBhY2tldCBlbmNvZGluZyBzY2hlbWUgZm9yIEJJRVIK ICAgdGhhdCBjYW4gaW1wcm92ZSB0aGUgZW5jb2RpbmcgZWZmaWNpZW5jeSBpbiBhIGxhcmdlIG5l dHdvcmsKICAgZXNwZWNpYWxseSB3aGVuIGdyb3VwIG1lbWJlcnMgYXJlIHNwYXJzZWx5IHBvcHVs YXRlZC4gIFRoZSBuZXcgc2NoZW1lCiAgIGFsc28gY2FuIGF2b2lkIHVubmVjZXNzYXJ5IHBhY2tl dCByZXBsaWNhdGlvbiBhdCBpbmdyZXNzIHdoZW4gbWVtYmVycwogICBvZiBhIHNwYXJzZWx5IHBv cHVsYXRlZCBtdWx0aWNhc3QgZ3JvdXAgc3BhbiBhY3Jvc3MgbXVsdGlwbGUKICAgZGlmZmVyZW50 IHJlZ2lvaW5zIChoYXZpbmcgZGlmZmVyZW50IFNJcykuCgo0LiAgSUFOQSBDb25zaWRlcmF0aW9u cwoKICAgVGhpcyBkb2N1bWVudCBkZWZpbmVzIG5vIG5ldyBuYW1lc3BhY2VzLgoKNS4gIFNlY3Vy aXR5IENvbnNpZGVyYXRpb25zCgogICBUaGlzIHByb3Bvc2FsIGJyb3VnaHQgbm8gbmV3IHNlY3Vy aXR5IGNvbmNlcm5zIG90aGVyIHRoYW4gdGhvc2UKICAgaWRlbnRpZmllZCBpbiBbSS1ELmRyYWZ0 LXdpam5hbmRzLWJpZXItYXJjaGl0ZWN0dXJlXS4KCjYuICBSZWZlcmVuY2VzCgo2LjEuICBOb3Jt YXRpdmUgUmVmZXJlbmNlcwoKICAgW1JGQzIxMTldICBCcmFkbmVyLCBTLiwgIktleSB3b3JkcyBm b3IgdXNlIGluIFJGQ3MgdG8gSW5kaWNhdGUKICAgICAgICAgICAgICBSZXF1aXJlbWVudCBMZXZl bHMiLCBCQ1AgMTQsIFJGQyAyMTE5LCBNYXJjaCAxOTk3LgoKNi4yLiAgSW5mb3JtYXRpdmUgUmVm ZXJlbmNlcwoKICAgW0ktRC53aWpuYW5kcy1iaWVyLWFyY2hpdGVjdHVyZV0KICAgICAgICAgICAg ICBXaWpuYW5kcywgSS4sIFJvc2VuLCBFLiwgRG9sZ2Fub3csIEEuLCBQcnp5Z2llbmRhLCBULiwg YW5kCiAgICAgICAgICAgICAgUy4gQWxkcmluLCAiTXVsdGljYXN0IHVzaW5nIEJpdCBJbmRleCBF eHBsaWNpdAogICAgICAgICAgICAgIFJlcGxpY2F0aW9uIiwgZHJhZnQtd2lqbmFuZHMtYmllci1h cmNoaXRlY3R1cmUtMDEgKHdvcmsgaW4KICAgICAgICAgICAgICBwcm9ncmVzcyksIE9jdG9iZXIg MjAxNC4KCiAgIFtJLUQud2lqbmFuZHMtbXBscy1iaWVyLWVuY2Fwc3VsYXRpb25dCiAgICAgICAg ICAgICAgV2lqbmFuZHMsIEkuLCBSb3NlbiwgRS4sIERvbGdhbm93LCBBLiwgVGFudHN1cmEsIEou LCBhbmQKICAgICAgICAgICAgICBTLiBBbGRyaW4sICJFbmNhcHN1bGF0aW9uIGZvciBCaXQgSW5k ZXggRXhwbGljaXQKICAgICAgICAgICAgICBSZXBsaWNhdGlvbiBpbiBNUExTIE5ldHdvcmtzIiwg ZHJhZnQtd2lqbmFuZHMtbXBscy1iaWVyLQogICAgICAgICAgICAgIGVuY2Fwc3VsYXRpb24tMDEg KHdvcmsgaW4gcHJvZ3Jlc3MpLCBPY3RvYmVyIDIwMTQuCgpBdXRob3JzJyBBZGRyZXNzZXMKCiAg IEFsYmVydCBUaWFuCiAgIEVyaWNzc29uLCBJbmMuCiAgIDI1MCBIb2xnZXIgV2F5LAogICBTYW4g Sm9zZSwgQ2FsaWZvcm5pYSAgOTUxMzQKICAgVVNBCgogICBFbWFpbDogYWxiZXJ0LnRpYW5AZXJp Y3Nzb24uY29tCgoKClRpYW4sIGV0IGFsLiAgICAgICAgICAgICAgIEV4cGlyZXMgTWF5IDksIDIw MTUgICAgICAgICAgICAgICAgICBbUGFnZSA1XQoMCkludGVybmV0LURyYWZ0ICAgICAgICBMaXN0 IEJhc2VkIEVuY29kaW5nIGZvciBCSUVSICAgICAgICAgTm92ZW1iZXIgMjAxNAoKCiAgIFJhbWFu YXRoYW4gTGFrc2htaWthbnRoYW4KICAgRXJpY3Nzb24sIEluYy4KICAgMjUwIEhvbGdlciBXYXks CiAgIFNhbiBKb3NlLCBDYWxpZm9ybmlhICA5NTEzNAogICBVU0EKCiAgIEVtYWlsOiByYW1hbmF0 aGFuLmxha3NobWlrYW50aGFuQGVyaWNzc29uLmNvbQoKCiAgIEplZmYgVGFudHN1cmEKICAgRXJp Y3Nzb24sIEluYy4KICAgMjAwIEhvbGdlciBXYXksCiAgIFNhbiBKb3NlLCBDYWxpZm9ybmlhICA5 NTEzNAogICBVU0EKCiAgIEVtYWlsOiBqZWZmLnRhbnRzdXJhQGVyaWNzc29uLmNvbQoKCiAgIEFu dG9uaSBQcnp5Z2llbmRhCiAgIEVyaWNzc29uLCBJbmMuCiAgIDMwMCBIb2xnZXIgV2F5LAogICBT YW4gSm9zZSwgQ2FsaWZvcm5pYSAgOTUxMzQKICAgVVNBCgogICBFbWFpbDogYW50b25pLnByenln aWVuZGFAZXJpY3Nzb24uY29tCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKCgoKVGlhbiwgZXQgYWwu ICAgICAgICAgICAgICAgRXhwaXJlcyBNYXkgOSwgMjAxNSAgICAgICAgICAgICAgICAgIFtQYWdl IDZdCg== --_005_D0812D8D311743AlbertTianericssoncom_-- From nobody Fri Nov 7 02:04:13 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D592C1ACF79 for ; Fri, 7 Nov 2014 02:04:10 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.794 X-Spam-Level: X-Spam-Status: No, score=-4.794 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ADq-J9xU9HBK for ; Fri, 7 Nov 2014 02:04:03 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 18D651ACF75 for ; Fri, 7 Nov 2014 02:04:02 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml401-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BLJ34299; Fri, 07 Nov 2014 10:04:01 +0000 (GMT) Received: from NKGEML401-HUB.china.huawei.com (10.98.56.32) by lhreml401-hub.china.huawei.com (10.201.5.240) with Microsoft SMTP Server (TLS) id 14.3.158.1; Fri, 7 Nov 2014 10:04:00 +0000 Received: from NKGEML512-MBS.china.huawei.com ([169.254.8.18]) by nkgeml401-hub.china.huawei.com ([10.98.56.32]) with mapi id 14.03.0158.001; Fri, 7 Nov 2014 18:03:55 +0800 From: Xuxiaohu To: "gjshep@gmail.com" , "bier@ietf.org" Thread-Topic: [Bier] BoF focus questions Thread-Index: AQHP7jm2pLxsEPDghkSTqhWhalqGVJxVA/UA Date: Fri, 7 Nov 2014 10:03:54 +0000 Message-ID: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE082CABBC@NKGEML512-MBS.china.huawei.com> References: In-Reply-To: Accept-Language: zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.98.134] Content-Type: multipart/alternative; boundary="_000_1FEE3F8F5CCDE64C9A8E8F4AD27C19EE082CABBCNKGEML512MBSchi_" MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/yD3-80t_WNwbhgTkBVxi-cgS_Ss Subject: Re: [Bier] BoF focus questions X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 07 Nov 2014 10:04:11 -0000 --_000_1FEE3F8F5CCDE64C9A8E8F4AD27C19EE082CABBCNKGEML512MBSchi_ Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 SGkgR3JlZywNCg0KQ2FuIHlvdSBhbGxvY2F0ZSBtZSBhIDEwLW1pbiBzbG90IHRvIHRhbGsgYWJv dXQgdGhlIGJpdHN0cmluZy1iYXNlZCBtdWx0aWNhc3QgZm9yd2FyZGluZyBwZXJmb3JtYW5jZSBh bmFseXNpcyAobm8gZHJhZnQsIG9ubHkgc2xpZGVzKT8gIEluIHRoaXMgcHJlc2VudGF0aW9uLCBJ IHdvdWxkIGV4cGxvcmUgcG9zc2libGUgYml0c3RyaW5nLWJhc2VkIG11bHRpY2FzdCBmb3J3YXJk aW5nIG1vZGVzIGFuZCB0aGVpciBjb3JyZXNwb25kaW5nIGlzc3Vlcy4gSSBiZWxpZXZlIHRoaXMg d291bGQgYmUgaGVscGZ1bCBmb3IgdXMgdG8gZXZhbHVhdGUgdGhlIGNvc3Qgb2Ygc3VjaCBuZXcg ZGF0YS1wbGFuZSBiZWhhdmlvci4NCg0KQmVzdCByZWdhcmRzLA0KWGlhb2h1DQoNCkZyb206IEJJ RVIgW21haWx0bzpiaWVyLWJvdW5jZXNAaWV0Zi5vcmddIE9uIEJlaGFsZiBPZiBHcmVnIFNoZXBo ZXJkDQpTZW50OiBUaHVyc2RheSwgT2N0b2JlciAyMywgMjAxNCA0OjUwIEFNDQpUbzogYmllckBp ZXRmLm9yZw0KU3ViamVjdDogW0JpZXJdIEJvRiBmb2N1cyBxdWVzdGlvbnMNCg0KKiwNCg0KRmly c3QsIHRoYW5rIGFsbCBvZiB5b3UgZm9yIGpvaW5pbmcgYW5kIGNvbnRyaWJ1dGluZyB0byBzb21l IGludGVyZXN0aW5nIGRpc2N1c3Npb24hIFRoaXMgaXMgcG90ZW50aWFsbHkgc29tZSBleGNpdGlu ZyBuZXcgd29yayBhbmQgdGhlIGRldGFpbGVkIGRpc2N1c3Npb25zIGhlcmUgb24gdGhlIGxpc3Qg d291bGQgaW5kaWNhdGUgdGhhdCB0aGVyZSBpcyBjb21tdW5pdHkgaW50ZXJlc3QgYWxyZWFkeS4N Cg0KQlVULCB0byBkYXRlIHRoZXJlIGlzIG5vIFdHLCB0aGVyZSBpcyBubyBhZG9wdGlvbiBvZiB0 aGlzIHdvcmsgYnkgYW55IG90aGVyIFdHLCBhbmQgd2UgaGF2ZW4ndCBldmVuIGhhZCBvdXIgZmly c3QgQm9GLiBQbGVhc2UgZmVlbCBmcmVlIHRvIGNvbnRpbnVlIHRoZSBiaXQtZmlkZGxpbmcgZGlz Y3Vzc2lvbnMgb24gdGhlIG90aGVyIHRocmVhZHMsIGJ1dCBhbHNvIHBsZWFzZSBqb2luIGluIGlu IGRpc2N1c3NpbmcgdGhlIGluaXRpYWwgcXVlc3Rpb25zIHdlIG5lZWQgdG8gYWRkcmVzcyBsZWFk aW5nIGludG8gYW5kIGR1cmluZyB0aGUgdXBjb21pbmcgQm9GOg0KDQoNCldoYXQgYXJlIHRoZSBw cm9ibGVtcyB0byBiZSBzb2x2ZWQ/DQoNCiAgIFBsZWFzZSBsb29rIGF0IHRoZSBjdXJyZW50IHBy b2JsZW0gc3RhdGVtZW50IGRyYWZ0IGFuZCB2b2ljZSB5b3VyIGFncmVlbWVudC9kaXNhZ3JlZW1l bnQvYWRkaXRpb25zIGFzIHlvdSBzZWUgdGhlbS4NCg0KaHR0cHM6Ly90b29scy5pZXRmLm9yZy9o dG1sL2RyYWZ0LXNoZXBoZXJkLWJpZXItcHJvYmxlbS1zdGF0ZW1lbnQtMDANCg0KDQoNCldoYXQg YXJlIHRoZSB1c2UgY2FzZXM/DQoNCiAgIFRoZXJlIGlzIGEgZHJhZnQgaW4gdGhlIHdvcmsgSSB1 bmRlcnN0YW5kLCBidXQgd2UgbmVlZCBpbnB1dCBmcm9tIHRoZSBicm9hZGVyIGNvbW11bml0eSBh Ym91dCBob3cgdGhpcyB3b3JrIG1heSBiZSBhYmxlIHRvIGFkZHJlc3Mgb3RoZXIgdXNlIGNhc2Vz IG5vdCB5ZXQgY29uc2lkZXJlZC4gSSBoYXZlIHJlY2VpdmVkIHNvbWUgcHJpdmF0ZSBlbWFpbCBh cm91bmQgdGhpcyBpc3N1ZSBhbmQgaW4gZXZlcnkgY2FzZSBhc2tlZCB0aGVtIHRvIHRha2UgaXQg dG8gdGhlIGxpc3QuIEhlcmUncyB5b3VyIGNoYW5jZS4NCg0KDQoNCldoYXQgaXMgdGhlIGNvc3Qg b2YgYSBuZXcgZGF0YS1wbGFuZSBiZWhhdmlvcj8NCg0KICAgVGhlIGN1cnJlbnQgYXJjaGl0ZWN0 dXJlIGRyYWZ0IHByZXNlbnRzIGEgc29sdXRpb24gd2hpY2ggcmVxdWlyZXMgYSBjb21wbGV0ZWx5 IG5ldyBkYXRhLXBsYW5lLiBUaGlzIG1heSBoYXZlIGEgY29uc2lkZXJhYmx5IGRpZmZlcmVudCBj b3N0IHRoYW4gYSBzb2x1dGlvbiB3aGljaCBjYW4gYmUgYWNoaWV2ZWQgYnkgc2ltcGx5IHR3ZWFr IGFuIGV4aXN0aW5nIHNvbHV0aW9uIG9yIGRhdGEtcGxhbmUuIFdoYXQgaXMgdGhlIG92ZXJhbGwg aW1wYWN0IG9mIGEgbmV3IGRhdGEtcGxhbmUgc29sdXRpb24/DQoNCmh0dHBzOi8vdG9vbHMuaWV0 Zi5vcmcvaHRtbC9kcmFmdC13aWpuYW5kcy1iaWVyLWFyY2hpdGVjdHVyZS0wMA0KDQoNCg0KRG9l cyB0aGlzIHByb2JsZW0gd2FycmFudCBhIG5ldyBkYXRhLXBsYW5lIGJlaGF2aW9yPw0KDQogICBL bm93aW5nIGFsbCB0aGUgYWJvdmUsIGRvIHdlIGZlZWwgdGhlIHBvdGVudGlhbCBiZW5lZml0cyBq dXN0aWZ5IHRoZSB3b3JrPw0KDQoNCg0KRG9uJ3QgaG9sZCBiYWNrLiBBbGwgaW5wdXQgaXMgdmFs aWQuIEl0J3MgYmVzdCB0aGF0IHdlIGJlZ2luIHRoaXMgZGlzY3Vzc2lvbiByaWdodCBhd2F5IHNv IHRoYXQgd2UgbWF5IG1vcmUgZWZmaWNpZW50bHkgZm9jdXMgdGhlIGRpc2N1c3Npb24gaW4gSGF3 YWlpLiBJJ20gbG9va2luZyBmb3J3YXJkIHRvIGhlYXJpbmcgZnJvbSB5b3UgYWxsIGhlcmUsIGFu ZCBzZWVpbmcgeW91IGluIEhJLiBZZXMsIHRoZXJlIHdpbGwgYmUgQmllciBhdCB0aGUgQklFUiBC b0YuIEhvdyBjb3VsZG4ndCB3ZT8NCg0KDQoNCkNoZWVycyEsDQoNCkdyZWcNCg0KDQoNCg0KDQoN Cg== --_000_1FEE3F8F5CCDE64C9A8E8F4AD27C19EE082CABBCNKGEML512MBSchi_ Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: base64 PGh0bWwgeG1sbnM6dj0idXJuOnNjaGVtYXMtbWljcm9zb2Z0LWNvbTp2bWwiIHhtbG5zOm89InVy bjpzY2hlbWFzLW1pY3Jvc29mdC1jb206b2ZmaWNlOm9mZmljZSIgeG1sbnM6dz0idXJuOnNjaGVt YXMtbWljcm9zb2Z0LWNvbTpvZmZpY2U6d29yZCIgeG1sbnM6bT0iaHR0cDovL3NjaGVtYXMubWlj cm9zb2Z0LmNvbS9vZmZpY2UvMjAwNC8xMi9vbW1sIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcv VFIvUkVDLWh0bWw0MCI+DQo8aGVhZD4NCjxtZXRhIGh0dHAtZXF1aXY9IkNvbnRlbnQtVHlwZSIg Y29udGVudD0idGV4dC9odG1sOyBjaGFyc2V0PXV0Zi04Ij4NCjxtZXRhIG5hbWU9IkdlbmVyYXRv ciIgY29udGVudD0iTWljcm9zb2Z0IFdvcmQgMTIgKGZpbHRlcmVkIG1lZGl1bSkiPg0KPHN0eWxl PjwhLS0NCi8qIEZvbnQgRGVmaW5pdGlvbnMgKi8NCkBmb250LWZhY2UNCgl7Zm9udC1mYW1pbHk6 SGVsdmV0aWNhOw0KCXBhbm9zZS0xOjIgMTEgNiA0IDIgMiAyIDIgMiA0O30NCkBmb250LWZhY2UN Cgl7Zm9udC1mYW1pbHk65a6L5L2TOw0KCXBhbm9zZS0xOjIgMSA2IDAgMyAxIDEgMSAxIDE7fQ0K QGZvbnQtZmFjZQ0KCXtmb250LWZhbWlseToiQ2FtYnJpYSBNYXRoIjsNCglwYW5vc2UtMToyIDQg NSAzIDUgNCA2IDMgMiA0O30NCkBmb250LWZhY2UNCgl7Zm9udC1mYW1pbHk6Q2FsaWJyaTsNCglw YW5vc2UtMToyIDE1IDUgMiAyIDIgNCAzIDIgNDt9DQpAZm9udC1mYWNlDQoJe2ZvbnQtZmFtaWx5 OlRhaG9tYTsNCglwYW5vc2UtMToyIDExIDYgNCAzIDUgNCA0IDIgNDt9DQpAZm9udC1mYWNlDQoJ e2ZvbnQtZmFtaWx5OiJcQOWui+S9kyI7DQoJcGFub3NlLTE6MiAxIDYgMCAzIDEgMSAxIDEgMTt9 DQovKiBTdHlsZSBEZWZpbml0aW9ucyAqLw0KcC5Nc29Ob3JtYWwsIGxpLk1zb05vcm1hbCwgZGl2 Lk1zb05vcm1hbA0KCXttYXJnaW46MGNtOw0KCW1hcmdpbi1ib3R0b206LjAwMDFwdDsNCglmb250 LXNpemU6MTIuMHB0Ow0KCWZvbnQtZmFtaWx5OuWui+S9kzt9DQphOmxpbmssIHNwYW4uTXNvSHlw ZXJsaW5rDQoJe21zby1zdHlsZS1wcmlvcml0eTo5OTsNCgljb2xvcjpibHVlOw0KCXRleHQtZGVj b3JhdGlvbjp1bmRlcmxpbmU7fQ0KYTp2aXNpdGVkLCBzcGFuLk1zb0h5cGVybGlua0ZvbGxvd2Vk DQoJe21zby1zdHlsZS1wcmlvcml0eTo5OTsNCgljb2xvcjpwdXJwbGU7DQoJdGV4dC1kZWNvcmF0 aW9uOnVuZGVybGluZTt9DQpwDQoJe21zby1zdHlsZS1wcmlvcml0eTo5OTsNCgltc28tbWFyZ2lu LXRvcC1hbHQ6YXV0bzsNCgltYXJnaW4tcmlnaHQ6MGNtOw0KCW1zby1tYXJnaW4tYm90dG9tLWFs dDphdXRvOw0KCW1hcmdpbi1sZWZ0OjBjbTsNCglmb250LXNpemU6MTIuMHB0Ow0KCWZvbnQtZmFt aWx5OuWui+S9kzt9DQpzcGFuLkVtYWlsU3R5bGUxOA0KCXttc28tc3R5bGUtdHlwZTpwZXJzb25h bC1yZXBseTsNCglmb250LWZhbWlseToiQ2FsaWJyaSIsInNhbnMtc2VyaWYiOw0KCWNvbG9yOiMx RjQ5N0Q7fQ0KLk1zb0NocERlZmF1bHQNCgl7bXNvLXN0eWxlLXR5cGU6ZXhwb3J0LW9ubHk7fQ0K QHBhZ2UgV29yZFNlY3Rpb24xDQoJe3NpemU6NjEyLjBwdCA3OTIuMHB0Ow0KCW1hcmdpbjo3Mi4w cHQgOTAuMHB0IDcyLjBwdCA5MC4wcHQ7fQ0KZGl2LldvcmRTZWN0aW9uMQ0KCXtwYWdlOldvcmRT ZWN0aW9uMTt9DQotLT48L3N0eWxlPjwhLS1baWYgZ3RlIG1zbyA5XT48eG1sPg0KPG86c2hhcGVk ZWZhdWx0cyB2OmV4dD0iZWRpdCIgc3BpZG1heD0iMTAyNiIgLz4NCjwveG1sPjwhW2VuZGlmXS0t PjwhLS1baWYgZ3RlIG1zbyA5XT48eG1sPg0KPG86c2hhcGVsYXlvdXQgdjpleHQ9ImVkaXQiPg0K PG86aWRtYXAgdjpleHQ9ImVkaXQiIGRhdGE9IjEiIC8+DQo8L286c2hhcGVsYXlvdXQ+PC94bWw+ PCFbZW5kaWZdLS0+DQo8L2hlYWQ+DQo8Ym9keSBsYW5nPSJaSC1DTiIgbGluaz0iYmx1ZSIgdmxp bms9InB1cnBsZSI+DQo8ZGl2IGNsYXNzPSJXb3JkU2VjdGlvbjEiPg0KPHAgY2xhc3M9Ik1zb05v cm1hbCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTYuMHB0O2ZvbnQtZmFt aWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0 OTdEIj5IaSBHcmVnLDxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi PjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjE2LjBwdDtmb250LWZhbWlseTom cXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+ PG86cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4g bGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTYuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0Nh bGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5DYW4geW91 IGFsbG9jYXRlIG1lIGEgMTAtbWluIHNsb3QgdG8gdGFsayBhYm91dCB0aGUgYml0c3RyaW5nLWJh c2VkIG11bHRpY2FzdCBmb3J3YXJkaW5nIHBlcmZvcm1hbmNlIGFuYWx5c2lzIChubyBkcmFmdCwg b25seSBzbGlkZXMpPyAmbmJzcDtJbiB0aGlzDQogcHJlc2VudGF0aW9uLCBJIHdvdWxkIGV4cGxv cmUgcG9zc2libGUgYml0c3RyaW5nLWJhc2VkIG11bHRpY2FzdCBmb3J3YXJkaW5nIG1vZGVzIGFu ZCB0aGVpciBjb3JyZXNwb25kaW5nIGlzc3Vlcy4gSSBiZWxpZXZlIHRoaXMgd291bGQgYmUgaGVs cGZ1bCBmb3IgdXMgdG8gZXZhbHVhdGUgdGhlIGNvc3Qgb2Ygc3VjaCBuZXcgZGF0YS1wbGFuZSBi ZWhhdmlvci4NCjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxz cGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjE2LjBwdDtmb250LWZhbWlseTomcXVv dDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+PG86 cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFu Zz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTYuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGli cmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5CZXN0IHJlZ2Fy ZHMsPG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFu Zz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTYuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGli cmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5YaWFvaHU8bzpw PjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5nPSJFTi1V UyIgc3R5bGU9ImZvbnQtc2l6ZToxNi4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90 OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPjxvOnA+Jm5ic3A7PC9vOnA+ PC9zcGFuPjwvcD4NCjxkaXYgc3R5bGU9ImJvcmRlcjpub25lO2JvcmRlci1sZWZ0OnNvbGlkIGJs dWUgMS41cHQ7cGFkZGluZzowY20gMGNtIDBjbSA0LjBwdCI+DQo8ZGl2Pg0KPGRpdiBzdHlsZT0i Ym9yZGVyOm5vbmU7Ym9yZGVyLXRvcDpzb2xpZCAjQjVDNERGIDEuMHB0O3BhZGRpbmc6My4wcHQg MGNtIDBjbSAwY20iPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PGI+PHNwYW4gbGFuZz0iRU4tVVMi IHN0eWxlPSJmb250LXNpemU6MTAuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O1RhaG9tYSZxdW90Oywm cXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij5Gcm9tOjwvc3Bhbj48L2I+PHNwYW4gbGFuZz0iRU4tVVMi IHN0eWxlPSJmb250LXNpemU6MTAuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O1RhaG9tYSZxdW90Oywm cXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij4gQklFUiBbbWFpbHRvOmJpZXItYm91bmNlc0BpZXRmLm9y Z10NCjxiPk9uIEJlaGFsZiBPZiA8L2I+R3JlZyBTaGVwaGVyZDxicj4NCjxiPlNlbnQ6PC9iPiBU aHVyc2RheSwgT2N0b2JlciAyMywgMjAxNCA0OjUwIEFNPGJyPg0KPGI+VG86PC9iPiBiaWVyQGll dGYub3JnPGJyPg0KPGI+U3ViamVjdDo8L2I+IFtCaWVyXSBCb0YgZm9jdXMgcXVlc3Rpb25zPG86 cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8L2Rpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi PjxzcGFuIGxhbmc9IkVOLVVTIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8ZGl2Pg0K PGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIGxhbmc9IkVOLVVTIj4qLDxvOnA+PC9v OnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFu IGxhbmc9IkVOLVVTIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxwIGNs YXNzPSJNc29Ob3JtYWwiPjxzcGFuIGxhbmc9IkVOLVVTIj5GaXJzdCwgdGhhbmsgYWxsIG9mIHlv dSBmb3Igam9pbmluZyBhbmQgY29udHJpYnV0aW5nIHRvIHNvbWUgaW50ZXJlc3RpbmcgZGlzY3Vz c2lvbiEgVGhpcyBpcyBwb3RlbnRpYWxseSBzb21lIGV4Y2l0aW5nIG5ldyB3b3JrIGFuZCB0aGUg ZGV0YWlsZWQgZGlzY3Vzc2lvbnMgaGVyZSBvbiB0aGUgbGlzdCB3b3VsZCBpbmRpY2F0ZSB0aGF0 IHRoZXJlIGlzIGNvbW11bml0eSBpbnRlcmVzdA0KIGFscmVhZHkuPG86cD48L286cD48L3NwYW4+ PC9wPg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIGxhbmc9IkVOLVVTIj48bzpw PiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9y bWFsIj48c3BhbiBsYW5nPSJFTi1VUyI+QlVULCB0byBkYXRlIHRoZXJlIGlzIG5vIFdHLCB0aGVy ZSBpcyBubyBhZG9wdGlvbiBvZiB0aGlzIHdvcmsgYnkgYW55IG90aGVyIFdHLCBhbmQgd2UgaGF2 ZW4ndCBldmVuIGhhZCBvdXIgZmlyc3QgQm9GLiBQbGVhc2UgZmVlbCBmcmVlIHRvIGNvbnRpbnVl IHRoZSBiaXQtZmlkZGxpbmcgZGlzY3Vzc2lvbnMgb24gdGhlIG90aGVyIHRocmVhZHMsIGJ1dCBh bHNvIHBsZWFzZSBqb2luDQogaW4gaW4gZGlzY3Vzc2luZyB0aGUgaW5pdGlhbCBxdWVzdGlvbnMg d2UgbmVlZCB0byBhZGRyZXNzIGxlYWRpbmcgaW50byBhbmQgZHVyaW5nIHRoZSB1cGNvbWluZyBC b0Y6PG86cD48L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05v cm1hbCI+PHNwYW4gbGFuZz0iRU4tVVMiPjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjwv ZGl2Pg0KPGRpdj4NCjxwIHN0eWxlPSJtYXJnaW46MGNtO21hcmdpbi1ib3R0b206LjAwMDFwdCI+ PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtIZWx2ZXRpY2EmcXVv dDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+V2hhdCBhcmUgdGhlIHByb2JsZW1zIHRvIGJlIHNv bHZlZD88bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBzdHlsZT0ibWFyZ2luOjBjbTttYXJnaW4t Ym90dG9tOi4wMDAxcHQiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1 b3Q7SGVsdmV0aWNhJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPiZuYnNwOyAmbmJzcDtQ bGVhc2UgbG9vayBhdCB0aGUgY3VycmVudCBwcm9ibGVtIHN0YXRlbWVudCBkcmFmdCBhbmQgdm9p Y2UgeW91ciBhZ3JlZW1lbnQvZGlzYWdyZWVtZW50L2FkZGl0aW9ucyBhcyB5b3Ugc2VlIHRoZW0u PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgc3R5bGU9Im1hcmdpbjowY207bWFyZ2luLWJvdHRv bTouMDAwMXB0Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0hl bHZldGljYSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij48YSBocmVmPSJodHRwczovL3Rv b2xzLmlldGYub3JnL2h0bWwvZHJhZnQtc2hlcGhlcmQtYmllci1wcm9ibGVtLXN0YXRlbWVudC0w MCI+aHR0cHM6Ly90b29scy5pZXRmLm9yZy9odG1sL2RyYWZ0LXNoZXBoZXJkLWJpZXItcHJvYmxl bS1zdGF0ZW1lbnQtMDA8L2E+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgc3R5bGU9Im1hcmdp bjowY207bWFyZ2luLWJvdHRvbTouMDAwMXB0Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZv bnQtZmFtaWx5OiZxdW90O0hlbHZldGljYSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij48 bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBzdHlsZT0ibWFyZ2luOjBjbTttYXJnaW4t Ym90dG9tOi4wMDAxcHQiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1 b3Q7SGVsdmV0aWNhJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPldoYXQgYXJlIHRoZSB1 c2UgY2FzZXM/PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgc3R5bGU9Im1hcmdpbjowY207bWFy Z2luLWJvdHRvbTouMDAwMXB0Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtZmFtaWx5 OiZxdW90O0hlbHZldGljYSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij4mbmJzcDsgJm5i c3A7VGhlcmUgaXMgYSBkcmFmdCBpbiB0aGUgd29yayBJIHVuZGVyc3RhbmQsIGJ1dCB3ZSBuZWVk IGlucHV0IGZyb20gdGhlIGJyb2FkZXIgY29tbXVuaXR5IGFib3V0IGhvdyB0aGlzIHdvcmsgbWF5 IGJlIGFibGUgdG8gYWRkcmVzcyBvdGhlciB1c2UgY2FzZXMNCiBub3QgeWV0IGNvbnNpZGVyZWQu IEkgaGF2ZSByZWNlaXZlZCBzb21lIHByaXZhdGUgZW1haWwgYXJvdW5kIHRoaXMgaXNzdWUgYW5k IGluIGV2ZXJ5IGNhc2UgYXNrZWQgdGhlbSB0byB0YWtlIGl0IHRvIHRoZSBsaXN0LiBIZXJlJ3Mg eW91ciBjaGFuY2UuPG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgc3R5bGU9Im1hcmdpbjowY207 bWFyZ2luLWJvdHRvbTouMDAwMXB0Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtZmFt aWx5OiZxdW90O0hlbHZldGljYSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij48bzpwPiZu YnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBzdHlsZT0ibWFyZ2luOjBjbTttYXJnaW4tYm90dG9t Oi4wMDAxcHQiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7SGVs dmV0aWNhJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPldoYXQgaXMgdGhlIGNvc3Qgb2Yg YSBuZXcgZGF0YS1wbGFuZSBiZWhhdmlvcj88bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBzdHls ZT0ibWFyZ2luOjBjbTttYXJnaW4tYm90dG9tOi4wMDAxcHQiPjxzcGFuIGxhbmc9IkVOLVVTIiBz dHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7SGVsdmV0aWNhJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYm cXVvdDsiPiZuYnNwOyAmbmJzcDtUaGUgY3VycmVudCBhcmNoaXRlY3R1cmUgZHJhZnQgcHJlc2Vu dHMgYSBzb2x1dGlvbiB3aGljaCByZXF1aXJlcyBhIGNvbXBsZXRlbHkgbmV3IGRhdGEtcGxhbmUu IFRoaXMgbWF5IGhhdmUgYSBjb25zaWRlcmFibHkgZGlmZmVyZW50IGNvc3QgdGhhbiBhDQogc29s dXRpb24gd2hpY2ggY2FuIGJlIGFjaGlldmVkIGJ5IHNpbXBseSB0d2VhayBhbiBleGlzdGluZyBz b2x1dGlvbiBvciBkYXRhLXBsYW5lLiBXaGF0IGlzIHRoZSBvdmVyYWxsIGltcGFjdCBvZiBhIG5l dyBkYXRhLXBsYW5lIHNvbHV0aW9uPzxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIHN0eWxlPSJt YXJnaW46MGNtO21hcmdpbi1ib3R0b206LjAwMDFwdCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxl PSJmb250LWZhbWlseTomcXVvdDtIZWx2ZXRpY2EmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90 OyI+PGEgaHJlZj0iaHR0cHM6Ly90b29scy5pZXRmLm9yZy9odG1sL2RyYWZ0LXdpam5hbmRzLWJp ZXItYXJjaGl0ZWN0dXJlLTAwIj5odHRwczovL3Rvb2xzLmlldGYub3JnL2h0bWwvZHJhZnQtd2lq bmFuZHMtYmllci1hcmNoaXRlY3R1cmUtMDA8L2E+PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAg c3R5bGU9Im1hcmdpbjowY207bWFyZ2luLWJvdHRvbTouMDAwMXB0Ij48c3BhbiBsYW5nPSJFTi1V UyIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0hlbHZldGljYSZxdW90OywmcXVvdDtzYW5zLXNl cmlmJnF1b3Q7Ij48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBzdHlsZT0ibWFyZ2lu OjBjbTttYXJnaW4tYm90dG9tOi4wMDAxcHQiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9u dC1mYW1pbHk6JnF1b3Q7SGVsdmV0aWNhJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPkRv ZXMgdGhpcyBwcm9ibGVtIHdhcnJhbnQgYSBuZXcgZGF0YS1wbGFuZSBiZWhhdmlvcj88bzpwPjwv bzpwPjwvc3Bhbj48L3A+DQo8cCBzdHlsZT0ibWFyZ2luOjBjbTttYXJnaW4tYm90dG9tOi4wMDAx cHQiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7SGVsdmV0aWNh JnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPiZuYnNwOyAmbmJzcDtLbm93aW5nIGFsbCB0 aGUgYWJvdmUsIGRvIHdlIGZlZWwgdGhlIHBvdGVudGlhbCBiZW5lZml0cyBqdXN0aWZ5IHRoZSB3 b3JrPzxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIHN0eWxlPSJtYXJnaW46MGNtO21hcmdpbi1i b3R0b206LjAwMDFwdCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LWZhbWlseTomcXVv dDtIZWx2ZXRpY2EmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+PG86cD4mbmJzcDs8L286 cD48L3NwYW4+PC9wPg0KPHAgc3R5bGU9Im1hcmdpbjowY207bWFyZ2luLWJvdHRvbTouMDAwMXB0 Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0hlbHZldGljYSZx dW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij5Eb24ndCBob2xkIGJhY2suIEFsbCBpbnB1dCBp cyB2YWxpZC4gSXQncyBiZXN0IHRoYXQgd2UgYmVnaW4gdGhpcyBkaXNjdXNzaW9uIHJpZ2h0IGF3 YXkgc28gdGhhdCB3ZSBtYXkgbW9yZSBlZmZpY2llbnRseSBmb2N1cyB0aGUgZGlzY3Vzc2lvbiBp biBIYXdhaWkuDQogSSdtIGxvb2tpbmcgZm9yd2FyZCB0byBoZWFyaW5nIGZyb20geW91IGFsbCBo ZXJlLCBhbmQgc2VlaW5nIHlvdSBpbiBISS4gWWVzLCB0aGVyZSB3aWxsIGJlIEJpZXIgYXQgdGhl IEJJRVIgQm9GLiBIb3cgY291bGRuJ3Qgd2U/PG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgc3R5 bGU9Im1hcmdpbjowY207bWFyZ2luLWJvdHRvbTouMDAwMXB0Ij48c3BhbiBsYW5nPSJFTi1VUyIg c3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0hlbHZldGljYSZxdW90OywmcXVvdDtzYW5zLXNlcmlm JnF1b3Q7Ij48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBzdHlsZT0ibWFyZ2luOjBj bTttYXJnaW4tYm90dG9tOi4wMDAxcHQiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1m YW1pbHk6JnF1b3Q7SGVsdmV0aWNhJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPkNoZWVy cyEsPG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgc3R5bGU9Im1hcmdpbjowY207bWFyZ2luLWJv dHRvbTouMDAwMXB0Ij48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90 O0hlbHZldGljYSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij5HcmVnPG86cD48L286cD48 L3NwYW4+PC9wPg0KPHAgc3R5bGU9Im1hcmdpbjowY207bWFyZ2luLWJvdHRvbTouMDAwMXB0Ij48 c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtZmFtaWx5OiZxdW90O0hlbHZldGljYSZxdW90 OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8 cCBzdHlsZT0ibWFyZ2luOjBjbTttYXJnaW4tYm90dG9tOi4wMDAxcHQiPjxzcGFuIGxhbmc9IkVO LVVTIiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7SGVsdmV0aWNhJnF1b3Q7LCZxdW90O3NhbnMt c2VyaWYmcXVvdDsiPjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjwvZGl2Pg0KPGRpdj4N CjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIGxhbmc9IkVOLVVTIj48bzpwPiZuYnNwOzwvbzpw Pjwvc3Bhbj48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBs YW5nPSJFTi1VUyI+PG86cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0KPC9kaXY+DQo8L2Rpdj4N CjwvZGl2Pg0KPC9kaXY+DQo8L2JvZHk+DQo8L2h0bWw+DQo= --_000_1FEE3F8F5CCDE64C9A8E8F4AD27C19EE082CABBCNKGEML512MBSchi_-- From nobody Sun Nov 9 09:19:34 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E3E491A1B9B for ; Sun, 9 Nov 2014 09:19:31 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.794 X-Spam-Level: X-Spam-Status: No, score=-4.794 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZTLwGaDduC97 for ; Sun, 9 Nov 2014 09:19:20 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5B8CD1A1BAC for ; Sun, 9 Nov 2014 09:19:20 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml404-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BLK97614; Sun, 09 Nov 2014 17:19:18 +0000 (GMT) Received: from DFWEML706-CHM.china.huawei.com (10.193.5.225) by lhreml404-hub.china.huawei.com (10.201.5.218) with Microsoft SMTP Server (TLS) id 14.3.158.1; Sun, 9 Nov 2014 17:19:18 +0000 Received: from DFWEML701-CHM.china.huawei.com ([10.193.5.50]) by dfweml706-chm ([10.193.5.225]) with mapi id 14.03.0158.001; Sun, 9 Nov 2014 09:19:08 -0800 From: Lucy yong To: "bier@ietf.org" , "gjshep@gmail.com" , "arkadiy.gulko@thomsonreuters.com" Thread-Topic: BIER Problem Statement Thread-Index: Ac/7puW3gS29625LRLK1ftGNYeM6UA== Date: Sun, 9 Nov 2014 17:19:08 +0000 Message-ID: <2691CE0099834E4A9C5044EEC662BB9D45409862@dfweml701-chm> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.47.150.183] Content-Type: multipart/alternative; boundary="_000_2691CE0099834E4A9C5044EEC662BB9D45409862dfweml701chm_" MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/55k_kqW97JEgjHR3Hu0qLfa59F8 Subject: [Bier] BIER Problem Statement X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 09 Nov 2014 17:19:32 -0000 --_000_2691CE0099834E4A9C5044EEC662BB9D45409862dfweml701chm_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Greg and Arkadiy, I just read BIER PS on air. This is the problem I see. Underlay IP network needs to support both unicast/multicast for overlay tra= nsport and need to reduce the cost and complex too. IGP network already has a lot of information that can be used for multicast= transport. Doing multicast in IGP network without PIM reduces # of protoco= ls, state, and multicast convergence time, which means a simpler underlay I= GP network that supports both unicast/multicast than the IGP+PIM network. Is this right understanding of the problem? Thanks, Lucy --_000_2691CE0099834E4A9C5044EEC662BB9D45409862dfweml701chm_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Hi Greg and Arkadiy,

 

I just read BIER PS on air.  This is the proble= m I see.

 

Underlay IP network needs to support both unicast/mu= lticast for overlay transport and need to reduce the cost and complex too.

 

IGP network already has a lot of information that ca= n be used for multicast transport. Doing multicast in IGP network without P= IM reduces # of protocols, state, and multicast convergence time, which mea= ns a simpler underlay IGP network that supports both unicast/multicast than the IGP+PIM network.

 

Is this right understanding of the problem?

 

Thanks,

Lucy

 

 

 

 

 

 

--_000_2691CE0099834E4A9C5044EEC662BB9D45409862dfweml701chm_-- From nobody Sun Nov 9 16:54:13 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3F29B1A87C3; Sun, 9 Nov 2014 16:54:11 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.55 X-Spam-Level: X-Spam-Status: No, score=-1.55 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_DE=0.35] autolearn=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id I4rnxM9EhN80; Sun, 9 Nov 2014 16:54:09 -0800 (PST) Received: from mailhost.informatik.uni-bremen.de (mailhost.informatik.uni-bremen.de [IPv6:2001:638:708:30c9::12]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5D1DB1A87DB; Sun, 9 Nov 2014 16:54:09 -0800 (PST) X-Virus-Scanned: amavisd-new at informatik.uni-bremen.de Received: from smtp-fb3.informatik.uni-bremen.de (smtp-fb3.informatik.uni-bremen.de [134.102.224.120]) by mailhost.informatik.uni-bremen.de (8.14.5/8.14.5) with ESMTP id sAA0rxXo014262; Mon, 10 Nov 2014 01:53:59 +0100 (CET) Received: from dhcp-a70a.meeting.ietf.org (dhcp-a70a.meeting.ietf.org [31.133.167.10]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by smtp-fb3.informatik.uni-bremen.de (Postfix) with ESMTPSA id F22FE5A0; Mon, 10 Nov 2014 01:53:57 +0100 (CET) Content-Type: text/plain; charset=windows-1252 Mime-Version: 1.0 (Mac OS X Mail 8.0 \(1990.1\)) From: Carsten Bormann In-Reply-To: <300194F7-D81F-4F2B-923A-BA6EF09DC334@tzi.org> Date: Sun, 9 Nov 2014 14:53:55 -1000 X-Mao-Original-Outgoing-Id: 437273635.15999-27f0abddbb6e36fc2cfe153fa71223e2 Content-Transfer-Encoding: quoted-printable Message-Id: References: <300194F7-D81F-4F2B-923A-BA6EF09DC334@tzi.org> To: IJsbrand Wijnands X-Mailer: Apple Mail (2.1990.1) Archived-At: http://mailarchive.ietf.org/arch/msg/bier/My_d1FfQODMHi7NzfCwgTeEQgKw Cc: Andrzej.Duda@imag.fr, roll@ietf.org, bier@ietf.org, 6lo@ietf.org Subject: Re: [Bier] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Nov 2014 00:54:11 -0000 On 04 Nov 2014, at 12:50, Carsten Bormann wrote: >=20 > It seems several people would be able to make a Sunday hallway chat = about this general approach. Since I=92m officially at a couple of = meetings (in parallel) from 1600, how about 1500? I=92ll be near the registration desk around 15:00 (in 5 minutes). Use my photo at http://www.ietf.org/wg/chair-photos.html to find me. Gr=FC=DFe, Carsten From nobody Sun Nov 9 22:43:56 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 607271A87BD for ; Sun, 9 Nov 2014 22:43:53 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.999 X-Spam-Level: X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yUtOhYk1OYe6 for ; Sun, 9 Nov 2014 22:43:50 -0800 (PST) Received: from mail-wg0-x233.google.com (mail-wg0-x233.google.com [IPv6:2a00:1450:400c:c00::233]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 2B6B01A892F for ; Sun, 9 Nov 2014 22:43:50 -0800 (PST) Received: by mail-wg0-f51.google.com with SMTP id l18so7911476wgh.38 for ; Sun, 09 Nov 2014 22:43:49 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:reply-to:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=bAtiOG36yPLir1RyOX1Zg74Xe9VBi6VZ8T7fXwr3d6c=; b=JHmZPQv18PT75b5UtA5UUDl11Nq2d+y2IbcXjf9KqnMwRH/HmSWp/Fs1kgYuPgQZGC PmNrrjTxdUoY/M0snEiDQ0zMzmRggIKaTI6OGXUEN5AMUyZ7nCV/9LBeauHWeBrS88jD ugInhSdKEM8RtBoJI595Eyns2Edpe7Xe0ygzz9vuabN35OWzDCJZa0AtwcqymIj7Of1G AubKaf834SkuyNWnRJ5fAlIGtjvhNdkUQe+yaaaVJL5SyBxRytFVBNcyfU55NJixs7u5 /3/Ip6EZ0mQ9zrKGRsxociP+xCEMuswuLYTYLof5BGR64vQ6POtdRdB3kSA0K4s8+B5S 2jpw== MIME-Version: 1.0 X-Received: by 10.180.10.105 with SMTP id h9mr13478758wib.36.1415601827273; Sun, 09 Nov 2014 22:43:47 -0800 (PST) Received: by 10.180.20.79 with HTTP; Sun, 9 Nov 2014 22:43:47 -0800 (PST) In-Reply-To: <2691CE0099834E4A9C5044EEC662BB9D45409862@dfweml701-chm> References: <2691CE0099834E4A9C5044EEC662BB9D45409862@dfweml701-chm> Date: Sun, 9 Nov 2014 22:43:47 -0800 Message-ID: From: Greg Shepherd To: Lucy yong Content-Type: multipart/alternative; boundary=001a11c263660ee45805077b7cd2 Archived-At: http://mailarchive.ietf.org/arch/msg/bier/RMNcL7JkU_recU3bAUr6ypgfjFA Cc: "arkadiy.gulko@thomsonreuters.com" , "bier@ietf.org" Subject: Re: [Bier] BIER Problem Statement X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gjshep@gmail.com List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Nov 2014 06:43:53 -0000 --001a11c263660ee45805077b7cd2 Content-Type: text/plain; charset=UTF-8 On Sun, Nov 9, 2014 at 9:19 AM, Lucy yong wrote: > Hi Greg and Arkadiy, > > > > I just read BIER PS on air. This is the problem I see. > > > > Underlay IP network needs to support both unicast/multicast for overlay > transport and need to reduce the cost and complex too. > > > > IGP network already has a lot of information that can be used for > multicast transport. Doing multicast in IGP network without PIM reduces # > of protocols, state, and multicast convergence time, which means a simpler > underlay IGP network that supports both unicast/multicast than the IGP+PIM > network. > > > > Is this right understanding of the problem? > Sorry, but no, this is not a restatement of the problem. Your text above is a loose description of a potential architecture. Greg Greg > Thanks, > > Lucy > > > > > > > > > > > > > --001a11c263660ee45805077b7cd2 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
On Sun, Nov 9, 2014 at 9:19 AM, Lucy yong <lucy.yong@h= uawei.com> wrote:

Hi Greg and Arkadiy,

=C2=A0

I just read BIER PS on air.=C2=A0 This is the proble= m I see.

=C2=A0

Underlay IP network needs to support both unicast/mu= lticast for overlay transport and need to reduce the cost and complex too.

=C2=A0

IGP network already has a lot of information that ca= n be used for multicast transport. Doing multicast in IGP network without P= IM reduces # of protocols, state, and multicast convergence time, which mea= ns a simpler underlay IGP network that supports both unicast/multicast than the IGP+PIM network.

=C2=A0

Is this right understanding of the problem?


Sorry, but no, this is not a restat= ement of the problem. Your text above is a loose description of a potential= architecture.

Greg

Greg<= /div>
=C2=A0

Thanks,

Lucy

=C2=A0

=C2=A0

=C2=A0

=C2=A0

=C2=A0

=C2=A0


--001a11c263660ee45805077b7cd2-- From nobody Mon Nov 10 09:03:16 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1BBC71A01D8; Mon, 10 Nov 2014 09:03:11 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.999 X-Spam-Level: X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id RlpOP6JrQBLo; Mon, 10 Nov 2014 09:03:09 -0800 (PST) Received: from mail-wi0-x230.google.com (mail-wi0-x230.google.com [IPv6:2a00:1450:400c:c05::230]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 0BFC31A01F6; Mon, 10 Nov 2014 09:03:06 -0800 (PST) Received: by mail-wi0-f176.google.com with SMTP id h11so11162666wiw.3 for ; Mon, 10 Nov 2014 09:03:03 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:reply-to:date:message-id:subject:from:to:content-type; bh=o2AAngpFFNGUv041OIZ0YLJU3UU4fWqZnrTXr0Zk010=; b=fdpZN5+w0xvud7rLPTznHYhmT1VrNSTvZR/HDLKpCbAiPAdwssXmmuiG+SkRauX4iF nJrI3eCY1H0zdwPgEyA5T2icYmpmcmakrbJw0rDP6ykw0ir2QUdlXVHEO5rWxdETdhmr hC+tQqxMsVoYRrVdgXauz/ntA6W5DCwZsNbzmePJUnRO4w3rJMIdVioDM3dzEBjEs8ce U2tD4XRv/K4oh0th4Bp/iczxa0cLIQXmgb7bBHKzUkiBELWWeUFxTME69MsyF0nopxjR nSQkD+8kNan/eGTy+3m+vIE9aL4YKz7/XtGf3iGmtnvB0z8ze2/e76N/8p/Px1rGaeFt 6lPg== MIME-Version: 1.0 X-Received: by 10.180.19.164 with SMTP id g4mr32042068wie.51.1415638983908; Mon, 10 Nov 2014 09:03:03 -0800 (PST) Received: by 10.180.20.79 with HTTP; Mon, 10 Nov 2014 09:03:03 -0800 (PST) Date: Mon, 10 Nov 2014 09:03:03 -0800 Message-ID: From: Greg Shepherd To: "bier@ietf.org" , "mboned@ietf.org" , "pim@ietf.org" , mpls@ietf.org, rtgwg@ietf.org Content-Type: multipart/alternative; boundary=bcaec53f386bc425bb050784220c Archived-At: http://mailarchive.ietf.org/arch/msg/bier/i_BWNIB5WZFZoaDHLNGJPtw-ekM Subject: [Bier] BIER BoF Today, w/ WebEx details X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gjshep@gmail.com List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Nov 2014 17:03:11 -0000 --bcaec53f386bc425bb050784220c Content-Type: text/plain; charset=UTF-8 bier Monday, November 10, 2014 1:00 pm | Hawaii Time (Honolulu, GMT-10:00) | 2 hr Join WebEx meeting: https://workgreen.webex.com/workgreen/j.php?MTID=mdfb183b4cd0d2cfaeceacf3950895a00 Meeting number: 825 308 148 Meeting password: 1234 Looking forward to seeing (hearing) you all there! Cheers, Greg --bcaec53f386bc425bb050784220c Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
bier
Monday, November 10= , 2014
1:00 pm | Hawaii Ti= me (Honolulu, GMT-10:00) | 2 hr

Join WebEx = meeting:
https://workgreen.webex.com/workgreen/j.php?MTID=3Dmdfb183= b4cd0d2cfaeceacf3950895a00
Me= eting number: 825 308 148
= Meeting password: 1234

Looking forward to seeing (hearing) you all there!

Cheers,
Greg
--bcaec53f386bc425bb050784220c-- From nobody Mon Nov 10 09:47:19 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ABF901A07BE; Mon, 10 Nov 2014 09:47:12 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.55 X-Spam-Level: X-Spam-Status: No, score=-1.55 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HELO_EQ_DE=0.35] autolearn=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id A6o6uXtflM2R; Mon, 10 Nov 2014 09:47:11 -0800 (PST) Received: from mailhost.informatik.uni-bremen.de (mailhost.informatik.uni-bremen.de [IPv6:2001:638:708:30c9::12]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 620881A0A6B; Mon, 10 Nov 2014 09:41:43 -0800 (PST) X-Virus-Scanned: amavisd-new at informatik.uni-bremen.de Received: from smtp-fb3.informatik.uni-bremen.de (smtp-fb3.informatik.uni-bremen.de [134.102.224.120]) by mailhost.informatik.uni-bremen.de (8.14.5/8.14.5) with ESMTP id sAAHfbEY029065; Mon, 10 Nov 2014 18:41:37 +0100 (CET) Received: from [192.168.1.100] (dhcp-939e.meeting.ietf.org [31.133.147.158]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by smtp-fb3.informatik.uni-bremen.de (Postfix) with ESMTPSA id 9ED60DAB; Mon, 10 Nov 2014 18:41:35 +0100 (CET) Content-Type: text/plain; charset=windows-1252 Mime-Version: 1.0 (Mac OS X Mail 8.0 \(1990.1\)) From: Carsten Bormann In-Reply-To: Date: Mon, 10 Nov 2014 07:41:32 -1000 X-Mao-Original-Outgoing-Id: 437334092.645113-d9cdd658dbe846e7d45b5a71c526c1e9 Content-Transfer-Encoding: quoted-printable Message-Id: <7D65A17E-75FF-4DD1-BF50-9AC7A7C03953@tzi.org> References: <300194F7-D81F-4F2B-923A-BA6EF09DC334@tzi.org> To: IJsbrand Wijnands X-Mailer: Apple Mail (2.1990.1) Archived-At: http://mailarchive.ietf.org/arch/msg/bier/PmTU0UOXeciXTey-v1t5tk9ivEM Cc: Andrzej.Duda@imag.fr, roll@ietf.org, bier@ietf.org, 6lo@ietf.org Subject: Re: [Bier] Source-Routed Multicast X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Nov 2014 17:47:12 -0000 With the I-D embargo lifted, I finally was able to post our I-D. Enjoy at: https://tools.ietf.org/html/draft-bergmann-bier-ccast If anybody wants me to talk about this at BIER, I also have slides. Gr=FC=DFe, Carsten From nobody Mon Nov 10 14:57:23 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BDB551ACE93; Mon, 10 Nov 2014 14:57:14 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.999 X-Spam-Level: X-Spam-Status: No, score=-3.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, GB_I_INVITATION=-2, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uS4LtaYx8PDu; Mon, 10 Nov 2014 14:57:12 -0800 (PST) Received: from mail-wg0-x232.google.com (mail-wg0-x232.google.com [IPv6:2a00:1450:400c:c00::232]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9CD0B1ACE28; Mon, 10 Nov 2014 14:57:12 -0800 (PST) Received: by mail-wg0-f50.google.com with SMTP id z12so10112518wgg.9 for ; Mon, 10 Nov 2014 14:57:11 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:reply-to:in-reply-to:references:date:message-id :subject:from:to:content-type; bh=4kQNRikaFlNuwBotqdgLn/990ofxi5muqrJBCTjzgNI=; b=NPpqq8pabeggPxa+q+HfHOWLEzS9XU60AkqVblpVuUX3foUaeACCjhW6aIkvUyfk92 OJVAoT/FBR2fG3ETTvdTptRXghpX8EL0z5kzfIE/Po1APxlJ/dnFYJPdM9rIoyRvJ8BP XqPP7IOQl/NB175K5YYpMucj3RbiE1a/vJrsVifG1tvuvtuWG3crS1VzVrrVUXBU3Kh6 81E0bNydV5r9mZ1b/aTYL8W2myvxE+tLBrPhDWDxuFxtu/liB4prNjCv8u6TyigzBiVR u0ObyBVW4jBqrXYF/NG6y6YpCPCnlaDjhE5+uehEcIu1ryHhxwZ8xwO/LGQpvTKVoONB KTkQ== MIME-Version: 1.0 X-Received: by 10.181.13.20 with SMTP id eu20mr34633679wid.36.1415660231372; Mon, 10 Nov 2014 14:57:11 -0800 (PST) Received: by 10.180.20.79 with HTTP; Mon, 10 Nov 2014 14:57:11 -0800 (PST) In-Reply-To: References: Date: Mon, 10 Nov 2014 14:57:11 -0800 Message-ID: From: Greg Shepherd To: "bier@ietf.org" , "mboned@ietf.org" , "pim@ietf.org" , mpls@ietf.org, rtgwg@ietf.org Content-Type: multipart/alternative; boundary=f46d043bdff836b6840507891565 Archived-At: http://mailarchive.ietf.org/arch/msg/bier/3-qayo9KNAgfAAzDSqAsN3h-YC8 Subject: Re: [Bier] BIER BoF Today, w/ WebEx details X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gjshep@gmail.com List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Nov 2014 22:57:15 -0000 --f46d043bdff836b6840507891565 Content-Type: text/plain; charset=UTF-8 Your WebEx session for IETF 91 has been scheduled. You should have just received an invitation in your email for this meeting; you should forward this information to your participants. Participants may also join from the http://www.ietf.org/meeting/91/remote-participation.html. Note that remote participants will need to listen to the audio stream (linked from both http://www.ietf.org/meeting/91/remote-participation.html and http://tools.ietf.org/agenda/91/) to follow the discussion. To claim the host role for your meeting, first join as a regular participant. Then, go to the participant menu at the top of the screen and choose "Claim Host Role." You will be prompted for the host key. Host Key: 684911 More information about WebEx may be found at http://www.ietf.org/wg/webex.html and http://www.ietf.org/wg/webex-hosting.pdf. Information about claiming the host role and running your WebEx meeting will also be found on the chairs' table at the front of the meeting room, and the host keys can be found at: http://www.ietf.org/private/webex-host-keys.html Best regards, Cindy On Mon, Nov 10, 2014 at 9:03 AM, Greg Shepherd wrote: > bier > Monday, November 10, 2014 > 1:00 pm | Hawaii Time (Honolulu, GMT-10:00) | 2 hr > > Join WebEx meeting: > > https://workgreen.webex.com/workgreen/j.php?MTID=mdfb183b4cd0d2cfaeceacf3950895a00 > Meeting number: 825 308 148 > Meeting password: 1234 > > Looking forward to seeing (hearing) you all there! > > Cheers, > Greg > --f46d043bdff836b6840507891565 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Your=C2=A0WebEx=C2=A0session for IETF 91 has been scheduled. You should have j= ust
received an invitation= in your email for this meeting; you should forward this
information to your participants. Participan= ts may also join from the
http://www.ietf.org/meeting/91/remote-participation.html.

Note that remote participants will need to listen to the audio st= ream (linked
from both=C2= =A0http://www.ietf.org/meeting/91/remote-participation.html=C2=A0and
http://tools.ietf.org/agenda/91/) to follow the discussion.

To claim the host role for your meeting, first join as a = regular participant.
Then,= go to the participant menu at the top of the screen and choose "Claim=
Host Role." You will= be prompted for the host key.

Host Key: 68= 4911

More information about=C2=A0WebEx=C2=A0may = be found at
http://www.ietf.org/wg/webex.html=C2=A0and
http://ww= w.ietf.org/wg/webex-hosting.pdf.

Information about claiming the host role and running your=C2=A0WebEx<= /span>=C2=A0mee= ting will
<= span style=3D"font-family:arial,sans-serif;font-size:13px">also be found on= the chairs' table at the front of the meeting room, and the
host keys can be found at:=C2=A0http://www.iet= f.org/private/webex-host-keys.html

Best regards,
Cindy

On Mon, Nov 10, 2014 at 9:03 AM, Greg Shepherd <= gjshep@gmail.com&= gt; wrote:
bier
Monday, November 10, 2014
1:00 pm | Hawaii Time (Honolulu, GMT-10:0= 0) | 2 hr
<= br style=3D"font-family:arial,sans-serif;font-size:13px">Join WebEx meeting:
https= ://workgreen.webex.com/workgreen/j.php?MTID=3Dmdfb183b4cd0d2cfaeceacf395089= 5a00
Meeting number: 825 308 = 148
Meeting password: 1234=

Looking forward to = seeing (hearing) you all there!

Cheers,
Greg

--f46d043bdff836b6840507891565-- From nobody Wed Nov 12 11:17:30 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EB1CE1A06E9 for ; Wed, 12 Nov 2014 11:17:28 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.999 X-Spam-Level: X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XMHDcAaHQVk3 for ; Wed, 12 Nov 2014 11:17:27 -0800 (PST) Received: from mail-wi0-x230.google.com (mail-wi0-x230.google.com [IPv6:2a00:1450:400c:c05::230]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D88DA1A0B00 for ; Wed, 12 Nov 2014 11:17:23 -0800 (PST) Received: by mail-wi0-f176.google.com with SMTP id h11so5971168wiw.3 for ; Wed, 12 Nov 2014 11:17:22 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:reply-to:date:message-id:subject:from:to:content-type; bh=DRSdg1AbGggLLWK/B+fWX9THPxoNJ9oXzL+tZPyYdBo=; b=d2/M3t/VcZ3bu1vtj1GAKTh0E8eyfa/G7BECS2DU9xlf5CKweWG2u3dSzfdiKRLXZF 99nYijL8Vh9NTcO93m8fz8cc24UcPZ9rSC3DdrFTIEXrwnTyE97WqfcdRIeAERs6bkic HkUVSp4xPQm6Xw1ZQFubc9HVwTA5LFy1XoBsn6xGSkunlXSXf/8kTXPxzbYhHJ6odEoR 1XamLwI1BaL4suAHaf7chE+l4NnHyEZxoPC0j19ivYrMe2wgOD49i3rGsl2mtAWDpmo9 9KTXSuLD6IClZWRemc6VbErvhKgReiqtB+33fHwP/C2dHp6XnAY1qc7BCtlhlXjOQ7Vd kfXQ== MIME-Version: 1.0 X-Received: by 10.180.87.196 with SMTP id ba4mr53335484wib.40.1415819842722; Wed, 12 Nov 2014 11:17:22 -0800 (PST) Received: by 10.180.20.79 with HTTP; Wed, 12 Nov 2014 11:17:22 -0800 (PST) Date: Wed, 12 Nov 2014 11:17:22 -0800 Message-ID: From: Greg Shepherd To: "bier@ietf.org" Content-Type: multipart/alternative; boundary=f46d0444ecb1caa20d0507ae3e3a Archived-At: http://mailarchive.ietf.org/arch/msg/bier/3ztcFhZUoBXT9lnH0OpJy8kDLxA Subject: [Bier] Minutes? X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gjshep@gmail.com List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 12 Nov 2014 19:17:29 -0000 --f46d0444ecb1caa20d0507ae3e3a Content-Type: text/plain; charset=UTF-8 Can the minute taker please send me the minutes please? Thanks, Greg --f46d0444ecb1caa20d0507ae3e3a Content-Type: text/html; charset=UTF-8
Can the minute taker please send me the minutes please?

Thanks,
Greg
--f46d0444ecb1caa20d0507ae3e3a-- From nobody Mon Nov 17 14:13:24 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4B2681ACD9A for ; Mon, 17 Nov 2014 14:13:18 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.794 X-Spam-Level: X-Spam-Status: No, score=-4.794 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Vk8j85nDlhSo for ; Mon, 17 Nov 2014 14:13:14 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E1E241ACDA1 for ; Mon, 17 Nov 2014 14:11:50 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml406-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BOX07574; Mon, 17 Nov 2014 22:11:48 +0000 (GMT) Received: from DFWEML704-CHM.china.huawei.com (10.193.5.141) by lhreml406-hub.china.huawei.com (10.201.5.243) with Microsoft SMTP Server (TLS) id 14.3.158.1; Mon, 17 Nov 2014 22:11:48 +0000 Received: from DFWEML703-CHM.china.huawei.com ([10.193.5.130]) by dfweml704-chm ([10.193.5.141]) with mapi id 14.03.0158.001; Mon, 17 Nov 2014 14:11:46 -0800 From: AshwoodsmithPeter To: "bier@ietf.org" Thread-Topic: Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZA== Date: Mon, 17 Nov 2014 22:11:45 +0000 Message-ID: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.193.60.150] Content-Type: multipart/alternative; boundary="_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208dfweml703chm_" MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/8rEpJzCPdaQWx4gy04hTaT3a1z4 Subject: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 17 Nov 2014 22:13:19 -0000 --_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208dfweml703chm_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable I made a couple of points in Honolulu about the inability of Bier to produ= ce optimum bandwidth trees (i.e Steiner) which I did not properly explain. = In particular the counter claim that Multi Topology (MT) could solve the pr= oblem is not correct. A Steiner tree is essentially a minimum cost tree that spans a subset of no= des in a graph. It will use the minimum bandwidth to deliver copies to all = members. If you add or remove a member the tree usually will be totally dif= ferent (i.e. more expensive, often considerably so). The computation of the Steiner tree is very hard, i.e. non-polynomial, in p= art because: - The Steiner tree for some set of nodes is NOT a subset of any tree comput= ed based on a superset of those nodes (i.e. not a subset of another Steiner= tree, or SPF tree, or spanning tree etc.). To compute the Steiner tree you therefore need the exact set of nodes to be= connected as input. Since Bier routing has no knowledge of the set of node= s being connected when it computes the path to a given node for a given MT = it is impossible for those computations to produce a proper Steiner tree fo= r each subset of nodes. The best that can be done with MT is to produce a small number of Steiner t= rees that span the entire set of edge nodes but since the subsets of those = trees are not Steiner we don't get Steiner trees between the actual members= of the bitset (which is usually a subset of what was used to compute the f= ull tree). Now does this matter? Well if you are trying to squeeze the last drop of b= andwidth out of the network you need a real Steiner tree for the set of nod= es being connected. Mind you if you are so concerned about bandwidth the 1= 0% or more overhead of the Bier header is likely also a concern. Now I'm not suggesting Bier is a bad idea, simply trying make sure people u= nderstand it cannot do what a statically configured P2MP tree could do in t= erms of bandwidth optimization. For other applications, such as PIM replac= ement it seems like a very interesting solution. Anyway a very interesting proposal and as far as creating a WG, I think tha= t if SPRING got a WG Bier deserves one. Cheers, Peter --_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208dfweml703chm_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

I made a couple of points in Honolulu about the inab= ility of Bier to  produce optimum bandwidth trees (i.e Steiner) which = I did not properly explain. In particular the counter claim that Multi Topo= logy (MT) could solve the problem is not correct.

 

A Steiner tree is essentially a minimum cost tree th= at spans a subset of nodes in a graph. It will use the minimum bandwidth to del= iver copies to all members. If you add or remove a member the tree usually = will be totally different (i.e. more expensive, often considerably so).

 

The computation of the Steiner tree is very hard, i.= e. non-polynomial, in part because:

- The Steiner tree for some set of nodes is NOT a su= bset of any tree computed based on a superset of those nodes (i.e. not a su= bset of another Steiner tree, or SPF tree, or spanning tree etc.).

 

To compute the Steiner tree you therefore need the e= xact set of nodes to be connected as input. Since Bier routing has no knowl= edge of the set of nodes being connected when it computes the path to a giv= en node for a given MT it is impossible for those computations to produce a proper Steiner tree for each subset of= nodes.

 

The best that can be done with MT is to produce a sm= all number of Steiner trees that span the entire set of edge nodes but since the subsets of those trees are no= t Steiner we don’t get Steiner trees between the actual members of th= e bitset (which is usually a subset of what was used to compute the full tr= ee).

 

Now does this matter? Well  if you are trying t= o squeeze the last drop of bandwidth out of the network you need a real Ste= iner tree for the set of nodes being connected.  Mind you if you are s= o concerned about bandwidth the 10% or more overhead of the Bier header is likely also a concern.

 

Now I’m not suggesting Bier is a bad idea, sim= ply trying make sure people understand it cannot do what a statically confi= gured P2MP tree could do in terms of bandwidth optimization.  For othe= r applications, such as PIM replacement it seems like a very interesting solution.

 

Anyway a very interesting proposal and as far as cre= ating a WG, I think that if SPRING got a WG Bier deserves one.

 

Cheers,

 

Peter

 

 

 

--_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208dfweml703chm_-- From nobody Mon Nov 17 14:44:55 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EF5861ACE3A for ; Mon, 17 Nov 2014 14:44:51 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.2 X-Spam-Level: X-Spam-Status: No, score=-4.2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5_rTqQMEuq7z for ; Mon, 17 Nov 2014 14:44:48 -0800 (PST) Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5D5051ACE3D for ; Mon, 17 Nov 2014 14:44:48 -0800 (PST) X-AuditID: c6180641-f79916d00000623a-7e-546a1ef49b4c Received: from EUSAAHC005.ericsson.se (Unknown_Domain [147.117.188.87]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id ED.FA.25146.4FE1A645; Mon, 17 Nov 2014 17:14:45 +0100 (CET) Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC005.ericsson.se ([147.117.188.87]) with mapi id 14.03.0174.001; Mon, 17 Nov 2014 17:44:43 -0500 From: Albert Tian To: AshwoodsmithPeter Thread-Topic: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AQHQArgUIN3t3ID8L06EJVKL7o34qg== Date: Mon, 17 Nov 2014 22:44:42 +0000 Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.4.140807 x-originating-ip: [147.117.188.9] Content-Type: multipart/alternative; boundary="_000_D08FB6393159DBAlbertTianericssoncom_" MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFvrBLMWRmVeSWpSXmKPExsUyuXRPuO5XuawQg54lEhZLZ+xhslh4t43J gcmj5chbVo8lS34yBTBFcdmkpOZklqUW6dslcGW8bjrNWPAiuGLTjSfsDYyb3boYOTkkBEwk Hh38yAxhi0lcuLeeDcQWEjjCKHH5C0sXIxeQvZxRYuqPL+wgCTYBHYlXsx8wgtgiAqYSK6f3 sIDYzALKEp83T2ICsYUFCiQmL/rBBFFTKPF9x2dmCFtP4sejNrBeFgFViU9/77KC2LwCZhKv fq8FW8wIdMT3U2uYIGaKS9x6Mp8J4jgBiSV7zkMdKirx8vE/sF5RoJnPNmxmh4grSuzrn84O 0Rst0bp8JRvEfEGJkzOfsExgFJmFZOwsJGWzkJRBxA0k3p+bzwxha0ssW/gaytaX2PjlLCOE bS3RO2MKG7KaBYwcqxg5SotTy3LTjQw3MQKj6pgEm+MOxgWfLA8xCnAwKvHwFuhkhgixJpYV V+YeYpTmYFES59WsnhcsJJCeWJKanZpakFoUX1Sak1p8iJGJg1OqgdH9OUv3Ev/jOUqRYovF VTRUxJTq3mdflEwpPLmFwYoxbzKj1Gcp146nXHFrrEJ0XY923zG3Dtz9x2Zl59JPR4P/3Al8 Ge5yb25o6E/L5TlT5x0s/24xuXJ7us6LoFPH9vzZIKj0a/c6w1Xu07p913dOOL1uwZnv2Tu8 dJok7G9cKNCd6ZknsVaJpTgj0VCLuag4EQBEgxy9iwIAAA== Archived-At: http://mailarchive.ietf.org/arch/msg/bier/JnY9dZb2od8gEKhnDs4KrD5GoUc Cc: "bier@ietf.org" Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 17 Nov 2014 22:44:52 -0000 --_000_D08FB6393159DBAlbertTianericssoncom_ Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable Hi Peter, One main factor of the BIER encoding efficiency comes from the fact that di= stributed BIFTs are already calculated on all the BFRs, and the BIFTs can b= e shared in the tree building exercise regardless of the group membership. So in the Steiner tree case, if the BIFT tables for each tree can not be sh= ared, then BIER may not be the right technology for the use case. However, using SPF, BIER can deliver the shortest path from each ingress to= each egress, which is a different optimization goal than Steiner=92s. Also= playing with the MT and BIFTs, BIER can support many tree types that can a= pproximate the Steiner. Thanks, Albert From: AshwoodsmithPeter > Date: Monday, November 17, 2014 at 2:11 PM To: "bier@ietf.org" > Subject: [Bier] Follow up on my comments from Honolulu concerning T.E/ Stei= ner Tree optimization with Bier I made a couple of points in Honolulu about the inability of Bier to produ= ce optimum bandwidth trees (i.e Steiner) which I did not properly explain. = In particular the counter claim that Multi Topology (MT) could solve the pr= oblem is not correct. A Steiner tree is essentially a minimum cost tree that spans a subset of no= des in a graph. It will use the minimum bandwidth to deliver copies to all = members. If you add or remove a member the tree usually will be totally dif= ferent (i.e. more expensive, often considerably so). The computation of the Steiner tree is very hard, i.e. non-polynomial, in p= art because: - The Steiner tree for some set of nodes is NOT a subset of any tree comput= ed based on a superset of those nodes (i.e. not a subset of another Steiner= tree, or SPF tree, or spanning tree etc.). To compute the Steiner tree you therefore need the exact set of nodes to be= connected as input. Since Bier routing has no knowledge of the set of node= s being connected when it computes the path to a given node for a given MT = it is impossible for those computations to produce a proper Steiner tree fo= r each subset of nodes. The best that can be done with MT is to produce a small number of Steiner t= rees that span the entire set of edge nodes but since the subsets of those = trees are not Steiner we don=92t get Steiner trees between the actual membe= rs of the bitset (which is usually a subset of what was used to compute the= full tree). Now does this matter? Well if you are trying to squeeze the last drop of b= andwidth out of the network you need a real Steiner tree for the set of nod= es being connected. Mind you if you are so concerned about bandwidth the 1= 0% or more overhead of the Bier header is likely also a concern. Now I=92m not suggesting Bier is a bad idea, simply trying make sure people= understand it cannot do what a statically configured P2MP tree could do in= terms of bandwidth optimization. For other applications, such as PIM repl= acement it seems like a very interesting solution. Anyway a very interesting proposal and as far as creating a WG, I think tha= t if SPRING got a WG Bier deserves one. Cheers, Peter --_000_D08FB6393159DBAlbertTianericssoncom_ Content-Type: text/html; charset="Windows-1252" Content-ID: Content-Transfer-Encoding: quoted-printable
Hi Peter,

One main factor of the BIER encoding efficiency comes from the fact th= at distributed BIFTs are already calculated on all the BFRs, and the BIFTs = can be shared in the tree building exercise regardless of the group members= hip.

So in the Steiner tree case, if the BIFT tables for each tree can not = be shared, then BIER may not be the right technology for the use case. = ;

However, using SPF, BIER can deliver the shortest path from each ingre= ss to each egress, which is a different optimization goal than Steiner=92s.= Also playing with the MT and BIFTs, BIER can support many tree types that = can approximate the Steiner.

Thanks,
Albert


From: AshwoodsmithPeter <Peter.AshwoodSmith@huawei.com= >
Date: Monday, November 17, 2014 at = 2:11 PM
To: "bier@ietf.org" <bier@i= etf.org>
Subject: [Bier] Follow up on my com= ments from Honolulu concerning T.E/ Steiner Tree optimization with Bier

I made a couple of points in Honolulu about the inab= ility of Bier to  produce optimum bandwidth trees (i.e Steiner) which = I did not properly explain. In particular the counter claim that Multi Topo= logy (MT) could solve the problem is not correct.

 

A Steiner tree is essentially a minimum cost tree th= at spans a subset of nodes in a graph. It will use the minimum bandwidth to del= iver copies to all members. If you add or remove a member the tree usually = will be totally different (i.e. more expensive, often considerably so).

 

The computation of the Steiner tree is very hard, i.= e. non-polynomial, in part because:

- The Steiner tree for some set of nodes is NOT a su= bset of any tree computed based on a superset of those nodes (i.e. not a su= bset of another Steiner tree, or SPF tree, or spanning tree etc.).

 

To compute the Steiner tree you therefore need the e= xact set of nodes to be connected as input. Since Bier routing has no knowl= edge of the set of nodes being connected when it computes the path to a giv= en node for a given MT it is impossible for those computations to produce a proper Steiner tree for each subset of= nodes.

 

The best that can be done with MT is to produce a sm= all number of Steiner trees that span the entire set of edge nodes but since the subsets of those trees are no= t Steiner we don=92t get Steiner trees between the actual members of the bi= tset (which is usually a subset of what was used to compute the full tree).=

 

Now does this matter? Well  if you are trying t= o squeeze the last drop of bandwidth out of the network you need a real Ste= iner tree for the set of nodes being connected.  Mind you if you are s= o concerned about bandwidth the 10% or more overhead of the Bier header is likely also a concern.

 

Now I=92m not suggesting Bier is a bad idea, simply = trying make sure people understand it cannot do what a statically configure= d P2MP tree could do in terms of bandwidth optimization.  For other ap= plications, such as PIM replacement it seems like a very interesting solution.

 

Anyway a very interesting proposal and as far as cre= ating a WG, I think that if SPRING got a WG Bier deserves one.

 

Cheers,

 

Peter

 

 

 

 --_000_D08FB6393159DBAlbertTianericssoncom_-- From nobody Mon Nov 17 15:10:58 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 48F881ACE69 for ; Mon, 17 Nov 2014 15:10:56 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.902 X-Spam-Level: X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JMz-w4GZGzzO for ; Mon, 17 Nov 2014 15:10:53 -0800 (PST) Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1bon0700.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::1:700]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 494021ACE70 for ; Mon, 17 Nov 2014 15:10:53 -0800 (PST) Received: from BY1PR0501MB1430.namprd05.prod.outlook.com (25.160.107.152) by BY1PR0501MB1430.namprd05.prod.outlook.com (25.160.107.152) with Microsoft SMTP Server (TLS) id 15.1.16.15; Mon, 17 Nov 2014 23:10:30 +0000 Received: from BY1PR0501MB1430.namprd05.prod.outlook.com ([25.160.107.152]) by BY1PR0501MB1430.namprd05.prod.outlook.com ([25.160.107.152]) with mapi id 15.01.0016.006; Mon, 17 Nov 2014 23:10:30 +0000 From: Ross Callon To: "bier@ietf.org" Thread-Topic: BIER and deterministic ECMP Thread-Index: AQHQAruuIBpZwkXRcE2ekC470Sj+dg== Date: Mon, 17 Nov 2014 23:10:30 +0000 Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [66.129.241.13] x-microsoft-antispam: BCL:0;PCL:0;RULEID:;SRVR:BY1PR0501MB1430; x-exchange-antispam-report-test: UriScan:; x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:; SRVR:BY1PR0501MB1430; x-forefront-prvs: 03982FDC1D x-forefront-antispam-report: SFV:NSPM; SFS:(10019020)(6009001)(189002)(164054003)(199003)(64706001)(86362001)(46102003)(20776003)(101416001)(66066001)(74316001)(2656002)(87936001)(54356999)(50986999)(92566001)(40100003)(122556002)(77096003)(62966003)(77156002)(450100001)(76576001)(99396003)(97736003)(2351001)(107046002)(108616004)(110136001)(95666004)(99286002)(21056001)(4396001)(31966008)(229853001)(33646002)(106356001)(106116001)(105586002)(120916001)(2501002)(24736002); DIR:OUT; SFP:1102; SCL:1; SRVR:BY1PR0501MB1430; H:BY1PR0501MB1430.namprd05.prod.outlook.com; FPR:; MLV:sfv; PTR:InfoNoRecords; MX:1; A:1; LANG:en; Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-OriginatorOrg: juniper.net Archived-At: http://mailarchive.ietf.org/arch/msg/bier/35ZQU_rtL8QKi19qg6hbPRj9FXM Cc: Ross Callon Subject: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 17 Nov 2014 23:10:56 -0000 Reading the BIER architecture (draft-wijnands-bier-architecture-01) I notic= ed the last sentence of section 6.7: Note however that by the rules of Section 6.5, any packet destined for both BFER-D and BFER-F will be sent via BFR-C. Thinking about this more: It is clear that where ECMP paths exist, which pa= th is taken to reach any particular destination will depend upon which other destinations the same packet goes to. In particular, which lower numbered bits are set in the BitString can change the choice among ECMP paths. This could have the effect that as a lower numbered destination intermittently joins and leaves a multicast group, the path to a higher numbered destination could switch back and forth between different ECMP paths. =20 This could be fixed if we used multiple BIFT tables, such that each BIFT ta= ble has only one path to each destination, but the multiple possible ECMP paths to any particular destination are spread across the multiple tables. When a BIER-encapsulated packet arrives to be forwarded, the BFR would use the BIE= R=20 Entropy field (or a hash thereon, if fewer than 256 next hops are supported= )=20 to determine which BIFT to use, and then the normal BIER forwarding algorit= hm=20 (the same as described in draft-wijnands-bier-architecture-01) would be use= d=20 with whichever BIFT table was selected. This would of course increase the=20 amount of memory that would be needed to store BIFT tables, but my quick "b= ack=20 of the envelope" calculation suggests that the amount of memory to store BI= FT=20 tables still should not be a problem if there aren't too many BIFT tables (= I=20 did my "back of the envelope" with max 16 tables). The forwarding algorithm= and=20 packet headers stay unchanged. This also has the feature that if this appro= ach=20 is consistently implemented across a service provider, then where a problem= has=20 been reported on a particular flow of packets to a particular receiver, we = can=20 traceroute with the same entropy value to that particular receiver and know= =20 that we will be using the same path as the normal data packets to that rece= iver. =20 I would be happy to write up appropriate text to add to draft-wijnands-bier= - architecture-01 (presumably for a -02 version) if the authors would like, o= r alternately could write up a separate Internet Draft to further describe th= e issue and the solution. Thanks, Ross From nobody Mon Nov 17 17:38:27 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 418B31AD03D for ; Mon, 17 Nov 2014 17:38:18 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.2 X-Spam-Level: X-Spam-Status: No, score=-4.2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5h_XyJiKvxr3 for ; Mon, 17 Nov 2014 17:38:09 -0800 (PST) Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id D009F1AD03A for ; Mon, 17 Nov 2014 17:38:08 -0800 (PST) X-AuditID: c6180641-f79916d00000623a-83-546a478f2097 Received: from EUSAAHC008.ericsson.se (Unknown_Domain [147.117.188.96]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id D3.7F.25146.0974A645; Mon, 17 Nov 2014 20:08:00 +0100 (CET) Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC008.ericsson.se ([147.117.188.96]) with mapi id 14.03.0174.001; Mon, 17 Nov 2014 20:38:00 -0500 From: Antoni Przygienda To: AshwoodsmithPeter , "bier@ietf.org" Thread-Topic: Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZAAG90tA Date: Tue, 18 Nov 2014 01:38:00 +0000 Message-ID: <2E4BB27CAB87BF43B4207C0E55860F181E75D1@eusaamb103.ericsson.se> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> In-Reply-To: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [147.117.188.9] Content-Type: multipart/alternative; boundary="_000_2E4BB27CAB87BF43B4207C0E55860F181E75D1eusaamb103ericsso_" MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrFLMWRmVeSWpSXmKPExsUyuXRPgu4E96wQg/2PZS2WztjDZLHwbhuT A5NHy5G3rB5LlvxkCmCK4rJJSc3JLEst0rdL4Mq42WtZMLmXsWLCzOusDYxHqroYOTkkBEwk pq1Ywgphi0lcuLeerYuRi0NI4AijxPWFj1ghnOWMEksXdrOBVLEJWEhc/vaUGcQWEYiQ6LnT zNjFyMEhLJAp8fmHH4gpIpAl8fCjG0SFkcTn3dOZQGwWAVWJpm3fwabwCnhLLOrcww5SLiTg InHnoStImFPAVaLn7imwcxiBzvl+ag1YK7OAuMStJ/OZIM4UkFiy5zwzhC0q8fLxP6jzFSX2 9U9nh6jPl5j65iwTxCpBiZMzn7BMYBSZhWTULCRls5CUQcR1JBbs/sQGYWtLLFv4mhnGPnPg MROy+AJG9lWMHKXFqWW56UaGmxiBkXNMgs1xB+OCT5aHGAU4GJV4eDe4ZoUIsSaWFVfmHmKU 5mBREufVrJ4XLCSQnliSmp2aWpBaFF9UmpNafIiRiYNTqoExofbUTV79upk9j+tfnW987X1k 7eLNNt+tLyg59wrebTFbu4nL/xS/X/PzpScmvvp5s+m66aTWmEV6x/P3vt38+cqXrb2bGrht wo5tWjhBusP6f771VWZN0W36b02DJgsmfcufw3IpYdPzkkVuH9V+KSTG7plv3loVEZ3vz5Uz h9n4zq8KNzNHJZbijERDLeai4kQALbv0aH0CAAA= Archived-At: http://mailarchive.ietf.org/arch/msg/bier/kD-f483GNqcOyDeTkri1LvugTvA Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 01:38:19 -0000 --_000_2E4BB27CAB87BF43B4207C0E55860F181E75D1eusaamb103ericsso_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable inline From: BIER [mailto:bier-bounces@ietf.org] On Behalf Of AshwoodsmithPeter Sent: Monday, November 17, 2014 2:12 PM To: bier@ietf.org Subject: [Bier] Follow up on my comments from Honolulu concerning T.E/ Stei= ner Tree optimization with Bier I made a couple of points in Honolulu about the inability of Bier to produ= ce optimum bandwidth trees (i.e Steiner) which I did not properly explain. = In particular the counter claim that Multi Topology (MT) could solve the pr= oblem is not correct. [Tony said] Well, no, MT per se cannot do that but you can restrict the t= opology using MT to be a Steiner itself so it's a 'poor man's' solution to = the problem ;-) A Steiner tree is essentially a minimum cost tree that spans a subset of no= des in a graph. It will use the minimum bandwidth to deliver copies to all = members. If you add or remove a member the tree usually will be totally dif= ferent (i.e. more expensive, often considerably so). The computation of the Steiner tree is very hard, i.e. non-polynomial, in p= art because: - The Steiner tree for some set of nodes is NOT a subset of any tree comput= ed based on a superset of those nodes (i.e. not a subset of another Steiner= tree, or SPF tree, or spanning tree etc.). [Tony said] yes. To compute the Steiner tree you therefore need the exact set of nodes to be= connected as input. Since Bier routing has no knowledge of the set of node= s being connected when it computes the path to a given node for a given MT = it is impossible for those computations to produce a proper Steiner tree fo= r each subset of nodes. [Tony said] It was never claimed that we compute for each subset of nodes, = a practical solution would be to compute a single steiner (MST) spanning _a= ll_ the BFRs with assigned ids (=3DBFERs & BFIRs) which should be still sig= nificantly better than SPFs in terms of replication (but that's all very ha= nd-wavy, pathological cases are easily constructed). The best that can be done with MT is to produce a small number of Steiner t= rees that span the entire set of edge nodes but since the subsets of those = trees are not Steiner we don't get Steiner trees between the actual members= of the bitset (which is usually a subset of what was used to compute the f= ull tree). [Tony said] yes Now does this matter? Well if you are trying to squeeze the last drop of b= andwidth out of the network you need a real Steiner tree for the set of nod= es being connected. Mind you if you are so concerned about bandwidth the 1= 0% or more overhead of the Bier header is likely also a concern. [Tony said] yes, but you don't provide any solution, just raise a problem. = Computing all the Steiners for all sets (which is combinatorial) is not fea= sible obviously. Now I'm not suggesting Bier is a bad idea, simply trying make sure people u= nderstand it cannot do what a statically configured P2MP tree could do in t= erms of bandwidth optimization. For other applications, such as PIM replac= ement it seems like a very interesting solution. [Tony said] A fully statically optimized solution based on known set of in= puts is always very hard to beat by a dynamic scheme. We can come up with = all kind of smarts such as measuring volumes between sets of BFIRs/BFERs an= d optimizing trees for those & pretty soon we'll have 32 'modes' of running= BIER. I think a single Steiner heuristics spanning all BFI/ERs would be interesti= ng enough. Optionally, we could stick a bit of BFRs saying "I'm preferred r= eplicator =3D core" of a tree and so on. Tons of work has been done in this= respect. n tony --_000_2E4BB27CAB87BF43B4207C0E55860F181E75D1eusaamb103ericsso_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

= inline

=  

From: BIER [mailto:bier-bounces@ietf.org] On Beha= lf Of AshwoodsmithPeter
Sent: Monday, November 17, 2= 014 2:12 PM
To: bier@ietf.org
Subject: [Bier] Follow up on= my comments from Honolulu concerning T.E/ Steiner Tree optimization with B= ier

 

I made a couple of points in Honolulu about the inability of= Bier to  produce optimum bandwidth trees (i.e Steiner) which I did no= t properly explain. In particular the counter claim that Multi Topology (MT) could solve the problem is not correct.

[Tony said]   Well, no, MT per se cannot do that but= you can restrict the topology using MT to be a Steiner itself so it’s a ‘poor man’s’ solution to the prob= lem ;-)

 

A Steiner tree is essentially a minimum cost tree that spans= a subset of nodes in a graph. It will use the minimum bandwidth to del= iver copies to all members. If you add or remove a member the tree usually = will be totally different (i.e. more expensive, often considerably so).

 

The computation of the Steiner tree is very hard, i.e. non-p= olynomial, in part because:

- The Steiner tree for some set of nodes is NOT a subset of = any tree computed based on a superset of those nodes (i.e. not a subset of = another Steiner tree, or SPF tree, or spanning tree etc.).

[Tony said] yes.

 

To compute the Steiner tree you therefore need the exact set= of nodes to be connected as input. Since Bier routing has no knowledge of = the set of nodes being connected when it computes the path to a given node for a given MT it is impossible for thos= e computations to produce a proper Steiner tree for each subset of nodes.

[Tony said] It was never claimed that we compute for each subs= et of nodes, a practical solution would be to compute a single steiner (MST) spanning _all_ the BFRs with assigned ids (=3DBFERs & BFIRs) which sh= ould be still significantly better than SPFs in terms of replication (but that’s all very hand-wavy, pathological= cases are easily constructed).

 

The best that can be done with MT is to produce a small numb= er of Steiner trees that span the entire set of edge nodes but since the subsets of those trees are no= t Steiner we don’t get Steiner trees between the actual members of th= e bitset (which is usually a subset of what was used to compute the full tr= ee).

[Tony said] yes<= span style=3D"color:#1F497D">

 

Now does this matter? Well  if you are trying to squeez= e the last drop of bandwidth out of the network you need a real Steiner tre= e for the set of nodes being connected.  Mind you if you are so concerned about bandwidth the 10% or more overhead of th= e Bier header is likely also a concern.

[Tony said] yes, but you don’t provide any solution, jus= t raise a problem. Computing all the Steiners for all sets (which is combinatorial) is not feasible obviously. <= /span>

 

Now I’m not suggesting Bier is a bad idea, simply tryi= ng make sure people understand it cannot do what a statically configured P2= MP tree could do in terms of bandwidth optimization.  For other applications, such as PIM replacement it seems like a very= interesting solution.

[Tony said]  A fully statically optimized solution based = on known set of inputs is always very hard to beat by a dynamic scheme.  We can come up with all kind of smarts such as = measuring volumes between sets of BFIRs/BFERs and optimizing trees for thos= e & pretty soon we’ll have 32 ‘modes’ of running BIER= .

 

I think a single Steiner heuristics spanning all BFI/ERs would= be interesting enough. Optionally, we could stick a bit of BFRs saying “I’m preferred replicator =3D core&= #8221; of a tree and so on. Tons of work has been done in this respect.

 

n  tony

 

--_000_2E4BB27CAB87BF43B4207C0E55860F181E75D1eusaamb103ericsso_-- From nobody Mon Nov 17 21:53:02 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 61B4A1A0102 for ; Mon, 17 Nov 2014 21:53:00 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.201 X-Spam-Level: X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Pr7ayoCet-Xz for ; Mon, 17 Nov 2014 21:52:57 -0800 (PST) Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id AB6171AD1C7 for ; Mon, 17 Nov 2014 21:52:57 -0800 (PST) X-AuditID: c618062d-f79206d0000014d2-64-546a85b46657 Received: from EUSAAHC004.ericsson.se (Unknown_Domain [147.117.188.84]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id 72.F8.05330.4B58A645; Tue, 18 Nov 2014 00:33:09 +0100 (CET) Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC004.ericsson.se ([147.117.188.84]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 00:52:56 -0500 From: Albert Tian To: Ross Callon , "bier@ietf.org" Thread-Topic: [Bier] BIER and deterministic ECMP Thread-Index: AQHQAvPmimRcG0ie5kmRBDyo9YL80Q== Date: Tue, 18 Nov 2014 05:52:55 +0000 Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.4.140807 x-originating-ip: [147.117.188.12] Content-Type: text/plain; charset="us-ascii" Content-ID: <8C7249C7067AA649926D2379A4FD8DF9@ericsson.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrALMWRmVeSWpSXmKPExsUyuXRPiO7W1qwQgxn/+S2WztjDZPF3xRUW ByaPJUt+Mnlcb7rKHsAUxWWTkpqTWZZapG+XwJVxaVcbS8E98Yp9b3+yNDCuF+5i5OSQEDCR eLP2GxuELSZx4d56IJuLQ0jgCKPEhIWP2CGc5YwSbz49ZgKpYhPQkXg1+wEjiC0i4Cbx8ewp sG5hAX2JfUd/AjVwAMUNJLqbuCBK9CQefG5iB7FZBFQlHt46xgpi8wqYSXz5vw7MZgRa/P3U GrDxzALiEreezGeCOEhAYsme88wQtqjEy8f/wOpFgWY+27CZHSKuJDHn9TVmiF4diQW7P7FB 2NYSG+9thLK1JZYtfM0MsVdQ4uTMJywTGEVnIVk3C0n7LCTts5C0z0LSvoCRdRUjR2lxallu upHBJkZglByTYNPdwbjnpeUhRgEORiUe3g2uWSFCrIllxZW5hxilOViUxHln1c4LFhJITyxJ zU5NLUgtii8qzUktPsTIxMEp1cC4XXjjxYXFsj8uhx73d1XZZPHx952rAZ/Me8MXGLs1TZ0/ aUH9h693fxX/XmjolfX9a/+fgKaCF8Z2qTGXd/Sa314193Nx8uG5LS/OfyvP15t/6tjq7Xqn 4nguK7y6nOB1p1Qr//4Fl+63bZpfdA6nF+ecaHYNZDQp3qit6s/6ut2mSCok7GCDEktxRqKh FnNRcSIAIeyYD3MCAAA= Archived-At: http://mailarchive.ietf.org/arch/msg/bier/x6Plebj3ivxfk7yUs7qQeFd6muk Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 05:53:00 -0000 Hi Ross, Very good point. Yes IMHO this needs to be addressed, otherwise the links leading to lower numbered BFERs will more likely be chosen in ECMP and cause unbalanced traffic. This maybe just as bad as the instability you have identified. Agreed multiple BIFT solution should work. Thanks, Albert On 11/17/14, 3:10 PM, "Ross Callon" wrote: >Reading the BIER architecture (draft-wijnands-bier-architecture-01) I >noticed >the last sentence of section 6.7: > > Note however that by the rules of Section 6.5, any packet destined > for both BFER-D and BFER-F will be sent via BFR-C. > >Thinking about this more: It is clear that where ECMP paths exist, which >path >is taken to reach any particular destination will depend upon which other >destinations the same packet goes to. In particular, which lower numbered >bits are set in the BitString can change the choice among ECMP paths. This >could have the effect that as a lower numbered destination intermittently >joins and leaves a multicast group, the path to a higher numbered >destination could switch back and forth between different ECMP paths. >=20 >This could be fixed if we used multiple BIFT tables, such that each BIFT >table >has only one path to each destination, but the multiple possible ECMP >paths >to any particular destination are spread across the multiple tables. When >a >BIER-encapsulated packet arrives to be forwarded, the BFR would use the >BIER=20 >Entropy field (or a hash thereon, if fewer than 256 next hops are >supported)=20 >to determine which BIFT to use, and then the normal BIER forwarding >algorithm=20 >(the same as described in draft-wijnands-bier-architecture-01) would be >used=20 >with whichever BIFT table was selected. This would of course increase the >amount of memory that would be needed to store BIFT tables, but my quick >"back=20 >of the envelope" calculation suggests that the amount of memory to store >BIFT=20 >tables still should not be a problem if there aren't too many BIFT tables >(I=20 >did my "back of the envelope" with max 16 tables). The forwarding >algorithm and=20 >packet headers stay unchanged. This also has the feature that if this >approach=20 >is consistently implemented across a service provider, then where a >problem has=20 >been reported on a particular flow of packets to a particular receiver, >we can=20 >traceroute with the same entropy value to that particular receiver and >know=20 >that we will be using the same path as the normal data packets to that >receiver. >=20 >I would be happy to write up appropriate text to add to >draft-wijnands-bier- >architecture-01 (presumably for a -02 version) if the authors would like, >or >alternately could write up a separate Internet Draft to further describe >the >issue and the solution. > >Thanks, Ross > > > >_______________________________________________ >BIER mailing list >BIER@ietf.org >https://www.ietf.org/mailman/listinfo/bier From nobody Tue Nov 18 07:44:11 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D06E01A1A8A for ; Tue, 18 Nov 2014 07:44:08 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.902 X-Spam-Level: X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id suJY4obMEV1b for ; Tue, 18 Nov 2014 07:44:00 -0800 (PST) Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0136.outbound.protection.outlook.com [65.55.169.136]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 697FD1A1A29 for ; Tue, 18 Nov 2014 07:44:00 -0800 (PST) Received: from [172.29.32.63] (66.129.241.11) by BY1PR0501MB1095.namprd05.prod.outlook.com (25.160.103.141) with Microsoft SMTP Server (TLS) id 15.1.16.15; Tue, 18 Nov 2014 15:43:57 +0000 Message-ID: <546B6938.9020709@juniper.net> Date: Tue, 18 Nov 2014 10:43:52 -0500 From: Eric C Rosen User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.2.0 MIME-Version: 1.0 To: AshwoodsmithPeter , "bier@ietf.org" References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> In-Reply-To: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: [66.129.241.11] X-ClientProxiedBy: BN3PR09CA0006.namprd09.prod.outlook.com (25.160.111.144) To BY1PR0501MB1095.namprd05.prod.outlook.com (25.160.103.141) X-Microsoft-Antispam: UriScan:; X-Microsoft-Antispam: BCL:0;PCL:0;RULEID:;SRVR:BY1PR0501MB1095; X-Exchange-Antispam-Report-Test: UriScan:; X-Exchange-Antispam-Report-CFA-Test: BCL:0; PCL:0; RULEID:; SRVR:BY1PR0501MB1095; X-Forefront-PRVS: 039975700A X-Forefront-Antispam-Report: SFV:NSPM; SFS:(10019020)(6049001)(6009001)(199003)(189002)(102836001)(101416001)(65956001)(106356001)(77156002)(47776003)(20776003)(33656002)(66066001)(40100003)(65806001)(86362001)(4396001)(92566001)(46102003)(92726001)(107046002)(64706001)(99396003)(105586002)(95666004)(122386002)(77096003)(59896002)(62966003)(97736003)(120916001)(36756003)(87976001)(64126003)(50466002)(42186005)(65816999)(2501002)(21056001)(31966008)(23746002)(83506001)(80316001)(87266999)(76176999)(54356999)(50986999); DIR:OUT; SFP:1102; SCL:1; SRVR:BY1PR0501MB1095; H:[172.29.32.63]; FPR:; MLV:sfv; PTR:InfoNoRecords; A:1; MX:1; LANG:en; X-Exchange-Antispam-Report-CFA-Test: BCL:0; PCL:0; RULEID:; SRVR:BY1PR0501MB1095; X-OriginatorOrg: juniper.net Archived-At: http://mailarchive.ietf.org/arch/msg/bier/76_NWNauJHUryehItexm_8yPlf8 Cc: erosen@juniper.net Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 15:44:10 -0000 > To compute the Steiner tree you therefore need the exact set of nodes > to be connected as input. Since Bier routing has no knowledge of the > set of nodes being connected when it computes the path to a given > node for a given MT it is impossible for those computations to > produce a proper Steiner tree for each subset of nodes. This is an interesting point, though I don't think it's stated quite right. A phrase like "BIER routing has no knowledge of the set of nodes being connected when it computes the path" is a bit imprecise, given the three-layer BIER architecture. In at least some applications, the multicast flow layer will know the exact set of nodes that needs to be connected for a given multicast flow. It is then possible for the multicast flow layer to cause a Steiner tree for that flow to be created. That tree can then be identified (e.g., assigned an MT-id), and used as an underlay by the BIER forwarding procedures. So I don't think there is anything in the BIER architecture that prevents the use of Steiner trees. However ... If you're going to build a tree that contains all and only the nodes through which a given flow needs to travel, you can use conventional multicast forwarding to send packets of that flow on that tree. BIER doesn't provide any added value in that case. Alternatively, as Tony suggests, one could build a Steiner tree connecting all the BFIRs/BFERs in the domain. Then one could send all flows along that tree, using BIER to prevent a given flow from going through nodes that it does not need to visit. I suspect Peter would say though that there is no point in doing this, as the path taken by any given flow would not be optimized (i.e., the path traveled by a given flow is not the path along the Steiner tree for that flow, and hence does not have the properties one might hope for). So I think the architecture does support the use case that Tony describes; whether that use case is actually useful or not is something we don't have to decide. From nobody Tue Nov 18 08:32:55 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CD76E1A1A7A for ; Tue, 18 Nov 2014 08:32:53 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.201 X-Spam-Level: X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id C80r2DsbdPQH for ; Tue, 18 Nov 2014 08:32:52 -0800 (PST) Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 068351A1A56 for ; Tue, 18 Nov 2014 08:32:51 -0800 (PST) X-AuditID: c618062d-f79206d0000014d2-3f-546b1ba82aa5 Received: from EUSAAHC001.ericsson.se (Unknown_Domain [147.117.188.75]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id AA.7C.05330.8AB1B645; Tue, 18 Nov 2014 11:12:56 +0100 (CET) Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC001.ericsson.se ([147.117.188.75]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 11:32:49 -0500 From: Antoni Przygienda To: Eric C Rosen , AshwoodsmithPeter , "bier@ietf.org" Thread-Topic: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZAAvONsAAAoVjxA= Date: Tue, 18 Nov 2014 16:32:48 +0000 Message-ID: <2E4BB27CAB87BF43B4207C0E55860F181E89BD@eusaamb103.ericsson.se> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> <546B6938.9020709@juniper.net> In-Reply-To: <546B6938.9020709@juniper.net> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [147.117.188.12] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFrrHLMWRmVeSWpSXmKPExsUyuXSPt+4K6ewQg1vXLC2WztjDZLFuwwdm i4V325gcmD1ajrxl9Viy5CeTx/Wmq+wBzFFcNimpOZllqUX6dglcGS/PvmYs6BWqOPBgF1sD 43K+LkZODgkBE4nzcy4zQdhiEhfurWfrYuTiEBI4wigxt6WVEcJZDuS8+8ICUsUmYCFx+dtT ZhBbRKBaYs/GT0DdHBzCAgUSS/arQIQLJb7v+AxVYiXx++t2MJtFQFVi396jbCA2r4C3xMqn sxhBbCGBVIkjr5+wgticAtoSPb07weKMQAd9P7UG7DhmAXGJW0/mQx0qILFkz3lmCFtU4uXj f6wQtpLEnNfXmCHqdSQW7P7EBmFrSyxb+JoZYq+gxMmZT1gmMIrOQjJ2FpKWWUhaZiFpWcDI soqRo7Q4tSw33chgEyMwQo5JsOnuYNzz0vIQowAHoxIP7wbXrBAh1sSy4srcQ4zSHCxK4ryz aucFCwmkJ5akZqemFqQWxReV5qQWH2Jk4uCUamCUEy0rkFc+Yrf4QdGCvxc+ON6IVj5/4SPj XK5d59ScObiTXaqurL63/fmLNb++tf/Zu+rBpP8Vmi1B932uvODPM9NYXyX50ObvFv/iBuP9 btMMeP2snjb/O32gwLpQen3pwQqLDVL/vdfHLg6NywsK4n856+fxVZkXpr9RL5Bdox5TGKhv kvRCiaU4I9FQi7moOBEA/wP9QXECAAA= Archived-At: http://mailarchive.ietf.org/arch/msg/bier/OblQkIBV3DXejiyLd66UYjPCx7o Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 16:32:54 -0000 > If you're going to build a tree that contains all and only the nodes thro= ugh > which a given flow needs to travel, you can use conventional multicast > forwarding to send packets of that flow on that tree. BIER doesn't provi= de > any added value in that case. >=20 > Alternatively, as Tony suggests, one could build a Steiner tree connectin= g all > the BFIRs/BFERs in the domain. Then one could send all flows along that > tree, using BIER to prevent a given flow from going through nodes that it > does not need to visit. I suspect Peter would say though that there is n= o > point in doing this, as the path taken by any given flow would not be > optimized (i.e., the path traveled by a given flow is not the path along = the > Steiner tree for that flow, and hence does not have the properties one mi= ght > hope for). >=20 > So I think the architecture does support the use case that Tony describes= ; > whether that use case is actually useful or not is something we don't hav= e to > decide. >=20 [Tony said] I would roughly agree with Eric here, if you have flows stable= and fat enough that you need to optimize distribution tree for specific re= ceiver/sender set and compute dedicated Steiner for it, amount of state in = the network becomes a secondary problem. Computer #1000s of Steiners for la= rge number of such combinations is not something I saw done since even the = heuristics on Steiner are quite beefy. I would still say that BIER provides an advantage due to minimizing the sta= te/protocols needed in the network if number of such trees is small [one tr= ee per sub-domain]. If number of such trees is large, it does not (since on= e would need a BIER sub-domain per receiver/sender set optimized by a Stein= er). Maybe BIER architecture can be easily extended to maintain set of com= puted trees in a sub-domain and choose the 'right' Steiner tree to deliver = traffic by setting it in the packet header but then one could argue, bitmas= k is not needed, the tree encodes the receivers/senders and compression of = such a thing quickly leads to (S,G) state in the core again is my intuition= [or alternately each tree represents a sub-domain but BIER is not made to = scale to huge numbers of sub-domains]=20 --- tony=20 From nobody Tue Nov 18 09:42:06 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 190621A1B5F for ; Tue, 18 Nov 2014 09:42:05 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.201 X-Spam-Level: X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id yJGK9Z01YFiK for ; Tue, 18 Nov 2014 09:42:02 -0800 (PST) Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id E70011A1B8C for ; Tue, 18 Nov 2014 09:41:54 -0800 (PST) X-AuditID: c618062d-f79206d0000014d2-4b-546b2bd638ce Received: from EUSAAHC006.ericsson.se (Unknown_Domain [147.117.188.90]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id 40.51.05330.6DB2B645; Tue, 18 Nov 2014 12:21:59 +0100 (CET) Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC006.ericsson.se ([147.117.188.90]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 12:41:51 -0500 From: Antoni Przygienda To: Ross Callon , "bier@ietf.org" Thread-Topic: BIER and deterministic ECMP Thread-Index: AQHQAruuIBpZwkXRcE2ekC470Sj+dpxmkeDw Date: Tue, 18 Nov 2014 17:41:50 +0000 Message-ID: <2E4BB27CAB87BF43B4207C0E55860F181E8A95@eusaamb103.ericsson.se> References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [147.117.188.12] Content-Type: multipart/mixed; boundary="_002_2E4BB27CAB87BF43B4207C0E55860F181E8A95eusaamb103ericsso_" MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFlrKIsWRmVeSWpSXmKPExsUyuXRPlO517ewQg/YbIhZLZ+xhsli34QOz xas5d5gt/q64wmKxb9E8RgdWjym/N7J6LFnyk8njetNV9gDmKC6blNSczLLUIn27BK6MJY9P Mxas163oPvmHvYHxsUYXIyeHhICJxMkFs9kgbDGJC/fWA9lcHEICRxglzn6czQzhLGeU2Drh GFgVm4CFxOVvT5lBbBEBN4mPZ0+BxZkFrjNKvNpgB2ILC2hIdByZzwhRoynR9HwaVL2RxK+5 y9m7GDk4WARUJT6eMQMJ8wp4S9xY+5QJxBYSiJBYPmE9K4jNKRApsehjMzuIzQh03PdTa5gg VolL3HoynwniaBGJhxdPQz0gKvHy8T9WCFtJYs7ra8wQ9ZkSq5ueskDsEpQ4OfMJywRG0VlI Rs1CUjYLSRlEXEdiwe5PbBC2tsSyha+ZIWwrieNNE1ghbE2Jvfs+A9lcQPZGRolFx56xQyQU JaZ0P2RfwMi5ipGjtDi1LDfdyGATIzBej0mw6e5g3PPS8hCjAAejEg/vBtesECHWxLLiytxD jNIcLErivLNq5wULCaQnlqRmp6YWpBbFF5XmpBYfYmTi4JRqYNxlP4NhBtfHQ4bRYnfYlTRW JDPuWvHT22RyekawR0N2xV+vzm2fQ54KaUR+mb1TfnHD05+GZV8ieS7JJ9+sum/cksAYdMm6 /IbioaIlAlq/d3mv4/s8OyVv8+LkaS1LzSPbP126vJdR6VtEVsLkmi3bbEp3bPryour3swm/ ym6rrnwv6btf9qgSS3FGoqEWc1FxIgALNa5KuAIAAA== Archived-At: http://mailarchive.ietf.org/arch/msg/bier/RQTzPMwccmiNL3aeeU-Ov9TFnvA Cc: Albert Tian , "Greg Shepherd \(shep\) \(shep@cisco.com\)" , "IJsbrand Wijnands \(ice@cisco.com\)" , "Eric Rosen \(erosen@juniper.net\)" Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 17:42:05 -0000 --_002_2E4BB27CAB87BF43B4207C0E55860F181E8A95eusaamb103ericsso_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable > Reading the BIER architecture (draft-wijnands-bier-architecture-01) I not= iced > the last sentence of section 6.7: >=20 > Note however that by the rules of Section 6.5, any packet destined > for both BFER-D and BFER-F will be sent via BFR-C. >=20 > Thinking about this more: It is clear that where ECMP paths exist, which = path > is taken to reach any particular destination will depend upon which other > destinations the same packet goes to. In particular, which lower numbered > bits are set in the BitString can change the choice among ECMP paths. Thi= s > could have the effect that as a lower numbered destination intermittently > joins and leaves a multicast group, the path to a higher numbered > destination could switch back and forth between different ECMP paths. [Tony said] Ross, I thought more about it & it's not a simple ECMP proble= m in my eyes. What you suggest [entry per destination] can lead to quite de= trimental effects under normal operation since we're operating here in the = domain of fan-out counts as well (I'll explain) [unless I misunderstood you= r email].=20 Observe the attached figure (I hope it survives the mailing list send for = others). When you're sending to a XYZ you clearly prefer the upper-most p= ath due 1 replication count (latest possible fan-out). Tables split per de= stination can lead to twice the bandwidth network consumption on the networ= k if A ends up replicating along ACX, ADY, AEZ (immediate fan-out). When X= falls off the receiver list, AD is the prefered 'axis' and if only Z is re= ceiver, you have 3 ECMP paths all consuming same amount of bandwidth.=20 My point: what is the property we try to preserve ? Given the example, I = would argue first that changing receiver list you may want as preference to= change the paths through the network since the optimal replication points = may shift to save bandwidth. Using ECMP too early can lead to unnecessary d= uplications. As long the receiver list is the same however, something lik= e entropy label should control the ECMP. =20 I am not saying the problem does not exist, I am just introducing another i= mportant axis that can deteriorate when looking @ optimizations in another = axis.=20 --- tony=20 --_002_2E4BB27CAB87BF43B4207C0E55860F181E8A95eusaamb103ericsso_ Content-Type: application/pdf; name="special_fishtail.pdf" Content-Description: special_fishtail.pdf Content-Disposition: attachment; filename="special_fishtail.pdf"; size=2114; creation-date="Tue, 18 Nov 2014 17:29:34 GMT"; modification-date="Tue, 18 Nov 2014 17:29:35 GMT" Content-Transfer-Encoding: base64 JVBERi0xLjQKJeLjz9MKNCAwIG9iago8PC9MZW5ndGggMTExNC9GaWx0ZXIvRmxhdGVEZWNvZGU+ PnN0cmVhbQp4nMVYTY/cNgy961f4mF4SiZRk+5i2aYACPWx3gH6hpwRJEDgF0h769yuSEiVt7ayL VWcw2PHjUtLT6NEUpc/m64vBOHn00+WteXUxd+azsZOf7fQpPe20mTDXJ/1/Mx/MT+YP46a/DUzf p1YfjbPTD+a33+30NnV/8gDnul+JBtFOzlI7HwVtjIAAORWIb9DkKq1yKGudiAxylz40vJ1+fG28 t8/DFNf0laZsyQg+fW3mvjXTuJEMXIuPUemH3HDfBzE9Z5t9dyNC5lz3Nf3+hZrRN/1tjBw3WRsg vkFKKKsyVNI6kR0lgBfYLbJsbiULXFnT1k5eYMsXL/DaS5v0e31Ua8cbuK+9gSLpJ07RSjAR2Bgg L00L4kA9lFMZsAWxV+PF63s3vf8rDUefP99nVYIsO/Izx7kYKMsozTojN/tg3j3U+Emj3dMcgeYo o6Y5pqT84rs035SU36WZ039d6k1B4CF9XT6ZZ6++unyUhH01tTFlHwwlFRKSXMQvADkViG9UKlRa 5VDWOpHDVLhmcdbAb0l5w1qbpGOBXJ/xnLU55Yl14GUa5/D67x/vDa4oQkiWJJY9SYH4Bm5OQqsc yloncqjI3G5ODnRVG7tsT27u13wpgsxkhF0fkm+9lRqrqrHqXj1riaBIvCP1WIseQlKJ62QOFYHQ SgLQFwxiqyRLL4n2Rc9b2Lzv5aoBQryNLBg1bdUoDSqGIhy5U1XiSlKJ62SOZenKOFj7NwXW9k1B eCDLbNuFn2Hfi/SewAK3kcXLbjIL2hihFtQV+aH7iRJXkkpcJ7MnC28T3hVZKAFhTVGNrbLUApuL O+3r6W3xdt73Iu/17gZvy0GRjbm0Ls//rcTGjvFEge3b+tppBd3aR/V1KOU1qaCpqfcF8sUb7CZr SR68GFFXQxVQSUbmLGVVhkpaJ3IoBkJ32ll6NWD5khqlLwTgWmzZ9/JZqGa7KypCaSEl5tQuZLQx QtnpOyTeQaJU4koSOtTS7R57cuZCWcQg2SbnJrEwHy+lZW+Vlt3ZZ9SQZw9Anit2D+UE9HM+Ae03 XjiDckZOje1za2FJLd6YZ9ZaR10tWdc9QZFMkumCL4mNEcvnezQyw1biShJ6tDwWQVKS5N0t5DIi yy2WzwJLy96a9yJo0JCnI4hPDBBLBP3yxQiaubHfjSBnbxhBci8YfLkFJBRL2CgYeh1ZaZUjtODh deRB9Kyt0msrdGyV7Yz1MHCeNNr5mKFEspSQ+fXxkIl7EZMW6EYRA3xtXC7pgIVCPRxCh2BkzFTi SoIdgseiJl+hloNOvvZsL9jc3F6qPbD2QmfUkOev7SgcnAbQt9e/t4OoF2h6XcYIiu4NGnnSqsSV BHsUTkUA+lYu9N0VK3R3rL3lDyPg6UP+twhALBHwzQ0iwNKJXypvRpsgKetbMI+Uv7Aqg+vR/Jj4 7SWI47OxWFuxypW6/ZcR95QfMd5Z2XkbkLMMqf6yUb18/gEXC8FmCmVuZHN0cmVhbQplbmRvYmoK NiAwIG9iago8PC9QYXJlbnQgNSAwIFIvQ29udGVudHMgNCAwIFIvVHlwZS9QYWdlL1Jlc291cmNl czw8L0V4dEdTdGF0ZTw8L0dTMiAyIDAgUi9HUzEgMSAwIFI+Pi9Qcm9jU2V0IFsvUERGIC9UZXh0 IC9JbWFnZUIgL0ltYWdlQyAvSW1hZ2VJXS9Gb250PDwvRjEgMyAwIFI+Pj4+L01lZGlhQm94WzAg MCA1NzAgNDcwXT4+CmVuZG9iagozIDAgb2JqCjw8L0Jhc2VGb250L0hlbHZldGljYS9UeXBlL0Zv bnQvRW5jb2RpbmcvV2luQW5zaUVuY29kaW5nL1N1YnR5cGUvVHlwZTE+PgplbmRvYmoKMSAwIG9i ago8PC9jYSAwPj4KZW5kb2JqCjIgMCBvYmoKPDwvY2EgMT4+CmVuZG9iago1IDAgb2JqCjw8L1R5 cGUvUGFnZXMvQ291bnQgMS9LaWRzWzYgMCBSXT4+CmVuZG9iago3IDAgb2JqCjw8L1R5cGUvQ2F0 YWxvZy9QYWdlcyA1IDAgUj4+CmVuZG9iago4IDAgb2JqCjw8L1Byb2R1Y2VyKGlUZXh0riA1LjQu MSCpMjAwMC0yMDEyIDFUM1hUIEJWQkEgXChBR1BMLXZlcnNpb25cKSkvTW9kRGF0ZShEOjIwMTQx MTE4MDkyOTM0LTA4JzAwJykvQ3JlYXRpb25EYXRlKEQ6MjAxNDExMTgwOTI5MzQtMDgnMDAnKT4+ CmVuZG9iagp4cmVmCjAgOQowMDAwMDAwMDAwIDY1NTM1IGYgCjAwMDAwMDE0NzYgMDAwMDAgbiAK MDAwMDAwMTUwMSAwMDAwMCBuIAowMDAwMDAxMzg4IDAwMDAwIG4gCjAwMDAwMDAwMTUgMDAwMDAg biAKMDAwMDAwMTUyNiAwMDAwMCBuIAowMDAwMDAxMTk3IDAwMDAwIG4gCjAwMDAwMDE1NzcgMDAw MDAgbiAKMDAwMDAwMTYyMiAwMDAwMCBuIAp0cmFpbGVyCjw8L1Jvb3QgNyAwIFIvSUQgWzxmYzBi ZmQyNDczYjMyMGExNmE2NzEzOGYwY2U5YjgyZT48ZmUyMjE1MWU2Y2YyYjlmOWZiOTk5MTIxYTQy OGVlNGE+XS9JbmZvIDggMCBSL1NpemUgOT4+CiVpVGV4dC01LjQuMQpzdGFydHhyZWYKMTc3NQol JUVPRgo= --_002_2E4BB27CAB87BF43B4207C0E55860F181E8A95eusaamb103ericsso_-- From nobody Tue Nov 18 10:46:30 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AE3971A6F03 for ; Tue, 18 Nov 2014 10:46:21 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.902 X-Spam-Level: X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rEkihGeHyF6j for ; Tue, 18 Nov 2014 10:46:17 -0800 (PST) Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1on0753.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::753]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 437891A6F2A for ; Tue, 18 Nov 2014 10:46:11 -0800 (PST) Received: from BY1PR0501MB1430.namprd05.prod.outlook.com (25.160.107.152) by BY1PR0501MB1095.namprd05.prod.outlook.com (25.160.103.141) with Microsoft SMTP Server (TLS) id 15.1.16.15; Tue, 18 Nov 2014 18:45:42 +0000 Received: from BY1PR0501MB1430.namprd05.prod.outlook.com ([25.160.107.152]) by BY1PR0501MB1430.namprd05.prod.outlook.com ([25.160.107.152]) with mapi id 15.01.0016.006; Tue, 18 Nov 2014 18:45:42 +0000 From: Ross Callon To: Antoni Przygienda , "bier@ietf.org" Thread-Topic: BIER and deterministic ECMP Thread-Index: AQHQAruuIBpZwkXRcE2ekC470Sj+dpxmkeDwgAAkxOA= Date: Tue, 18 Nov 2014 18:45:42 +0000 Message-ID: <3f8182b06bf741e3ad3e750db7ded76a@BY1PR0501MB1430.namprd05.prod.outlook.com> References: <2E4BB27CAB87BF43B4207C0E55860F181E8A95@eusaamb103.ericsson.se> In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F181E8A95@eusaamb103.ericsson.se> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [66.129.241.13] x-microsoft-antispam: BCL:0;PCL:0;RULEID:;SRVR:BY1PR0501MB1095; x-exchange-antispam-report-test: UriScan:; x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:; SRVR:BY1PR0501MB1095; x-forefront-prvs: 039975700A x-forefront-antispam-report: SFV:NSPM; SFS:(10019020)(6009001)(13464003)(199003)(51704005)(189002)(377454003)(101416001)(106356001)(106116001)(77156002)(76576001)(20776003)(66066001)(40100003)(92566001)(86362001)(4396001)(46102003)(64706001)(107046002)(108616004)(99396003)(105586002)(99286002)(95666004)(77096003)(122556002)(62966003)(97736003)(2656002)(120916001)(74316001)(19580395003)(2501002)(21056001)(31966008)(19580405001)(33646002)(87936001)(76176999)(54356999)(50986999)(24736002); DIR:OUT; SFP:1102; SCL:1; SRVR:BY1PR0501MB1095; H:BY1PR0501MB1430.namprd05.prod.outlook.com; FPR:; MLV:sfv; PTR:InfoNoRecords; A:1; MX:1; LANG:en; Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-OriginatorOrg: juniper.net Archived-At: http://mailarchive.ietf.org/arch/msg/bier/UD78i3GDaREDomG2lDx7qCHt9XM Cc: Ross Callon , Albert Tian , "Greg Shepherd \(shep\) \(shep@cisco.com\)" , "IJsbrand Wijnands \(ice@cisco.com\)" , Eric Rosen Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 18:46:23 -0000 Yes, there is a tradeoff. Suppose in your figure that you remove the link from C to Z. In that case, = a packet destined for Y and Z would be more efficiently sent to D (since on= ly one copy needs to be sent by A to D, and D can replicate to Y and Z. How= ever, a packet destined for X and Y would more efficiently be sent to C, si= nce A would send only one copy (to C) and C can replicate to X and Y. Suppose that X and Z kept joining and leaving the group, so that some packe= ts go only to X and Y, and some packets go only to Y and Z. In this case th= ere is a legitimate question: What behavior do we want? If we minimize traf= fic on the links out of A then the packets to X and Y would go via C, and t= he packets to Y and Z would go via D. This would mean that the packets to Y= keep fluctuating between paths (or at least fluctuate as quickly as X and = Z join and leave the group). If we wanted steady performance and predictabi= lity for the traffic to Y, we might prefer to have the traffic to Y always = take the same path at the cost of sometimes having to send two packets out = of node A.=20 Ross PS: Your figure showed up and printed fine, at least in my email system. =20 -----Original Message----- From: Antoni Przygienda [mailto:antoni.przygienda@ericsson.com]=20 Sent: Tuesday, November 18, 2014 12:42 PM To: Ross Callon; bier@ietf.org Cc: IJsbrand Wijnands (ice@cisco.com); Albert Tian; Eric Rosen; Greg Shephe= rd (shep) (shep@cisco.com) Subject: RE: BIER and deterministic ECMP > Reading the BIER architecture (draft-wijnands-bier-architecture-01) I not= iced > the last sentence of section 6.7: >=20 > Note however that by the rules of Section 6.5, any packet destined > for both BFER-D and BFER-F will be sent via BFR-C. >=20 > Thinking about this more: It is clear that where ECMP paths exist, which = path > is taken to reach any particular destination will depend upon which other > destinations the same packet goes to. In particular, which lower numbered > bits are set in the BitString can change the choice among ECMP paths. Thi= s > could have the effect that as a lower numbered destination intermittently > joins and leaves a multicast group, the path to a higher numbered > destination could switch back and forth between different ECMP paths. [Tony said] Ross, I thought more about it & it's not a simple ECMP proble= m in my eyes. What you suggest [entry per destination] can lead to quite de= trimental effects under normal operation since we're operating here in the = domain of fan-out counts as well (I'll explain) [unless I misunderstood you= r email].=20 Observe the attached figure (I hope it survives the mailing list send for = others). When you're sending to a XYZ you clearly prefer the upper-most p= ath due 1 replication count (latest possible fan-out). Tables split per de= stination can lead to twice the bandwidth network consumption on the networ= k if A ends up replicating along ACX, ADY, AEZ (immediate fan-out). When X= falls off the receiver list, AD is the prefered 'axis' and if only Z is re= ceiver, you have 3 ECMP paths all consuming same amount of bandwidth.=20 My point: what is the property we try to preserve ? Given the example, I = would argue first that changing receiver list you may want as preference to= change the paths through the network since the optimal replication points = may shift to save bandwidth. Using ECMP too early can lead to unnecessary d= uplications. As long the receiver list is the same however, something lik= e entropy label should control the ECMP. =20 I am not saying the problem does not exist, I am just introducing another i= mportant axis that can deteriorate when looking @ optimizations in another = axis.=20 --- tony=20 From nobody Tue Nov 18 10:50:58 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 17E621A6F3B for ; Tue, 18 Nov 2014 10:50:56 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.201 X-Spam-Level: X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wPp1uEBw4baG for ; Tue, 18 Nov 2014 10:50:54 -0800 (PST) Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id F0BF31A6F38 for ; Tue, 18 Nov 2014 10:50:53 -0800 (PST) X-AuditID: c6180641-f79916d00000623a-28-546b399817c0 Received: from EUSAAHC008.ericsson.se (Unknown_Domain [147.117.188.96]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id E3.AF.25146.8993B645; Tue, 18 Nov 2014 13:20:41 +0100 (CET) Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC008.ericsson.se ([147.117.188.96]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 13:50:50 -0500 From: Antoni Przygienda To: Ross Callon , "bier@ietf.org" Thread-Topic: BIER and deterministic ECMP Thread-Index: AQHQAruuIBpZwkXRcE2ekC470Sj+dpxmkeDwgAAkxOCAAAPQkA== Date: Tue, 18 Nov 2014 18:50:49 +0000 Message-ID: <2E4BB27CAB87BF43B4207C0E55860F181E8B1A@eusaamb103.ericsson.se> References: <2E4BB27CAB87BF43B4207C0E55860F181E8A95@eusaamb103.ericsson.se> <3f8182b06bf741e3ad3e750db7ded76a@BY1PR0501MB1430.namprd05.prod.outlook.com> In-Reply-To: <3f8182b06bf741e3ad3e750db7ded76a@BY1PR0501MB1430.namprd05.prod.outlook.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [147.117.188.12] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFvrGLMWRmVeSWpSXmKPExsUyuXRPgu5My+wQg5P3hCyWztjDZLFuwwdm i1dz7jBb/F1xhcVi36J5jA6sHlN+b2T1WLLkJ5PH9aar7AHMUVw2Kak5mWWpRfp2CVwZK99r FkwUqDgy+wFTA+MZni5GTg4JAROJGd9/MELYYhIX7q1n62Lk4hASOMIo0dOzCcpZziixa/Zr JpAqNgELicvfnjKD2CICbhIfz54CK2IW2McoMf/xEbAiYQENiY4j8xkhijQlmp5Pg2pwkti3 oZ8dxGYRUJX4+3gCWA2vgLfEl0vnWSC2vWSU2PF+ClgDp0CkxPbPW1hBbEag+76fWgO2gFlA XOLWk/lMEHcLSCzZc54ZwhaVePn4HyuErSQx5/U1Zoh6HYkFuz+xQdjaEssWvmaGWCwocXLm E5YJjGKzkIydhaRlFpKWWUhaFjCyrGLkKC1OLctNNzLcxAiMqGMSbI47GBd8sjzEKMDBqMTD u8E1K0SINbGsuDL3EKM0B4uSOK9m9bxgIYH0xJLU7NTUgtSi+KLSnNTiQ4xMHJxSDYw+684L CW/185U7y1ouy8x/TvUqb86uzEPa95r5zl+dLBzMrRSw1bP82DOLYqV7bJkXuBkkHwlHXSnc 91BlsdjFaS8mSHW+aWmW/NVtVe3Q2LnpnRYP96xAzePtLhMK2P+84Szrcy2sVe48ojZpc2h1 7se84OJLbWEaKctC97yfI3H177UFqkosxRmJhlrMRcWJAHWDBSqJAgAA Archived-At: http://mailarchive.ietf.org/arch/msg/bier/VuXt4MaAYeiFBFQxQN5ReyNaXzY Cc: Albert Tian , "Greg Shepherd \(shep\) \(shep@cisco.com\)" , "IJsbrand Wijnands \(ice@cisco.com\)" , Eric Rosen Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 18:50:56 -0000 Yepp, we're in sync. I think it's pretty much a vendor optimization we're = talking here but merits a draft explaining the trade-off space. The only p= roperty I see as uniformly desirable is that while the receiver set and lab= el stack is stable, OAM must be deterministic as to which of the ECMP paths= it's following (i.e. predictable ECMP, probably based on entropy label if = I'm not mistaken but I stand corrected otherwise). --- tony=20 > -----Original Message----- > From: Ross Callon [mailto:rcallon@juniper.net] > Sent: Tuesday, November 18, 2014 10:46 AM > To: Antoni Przygienda; bier@ietf.org > Cc: IJsbrand Wijnands (ice@cisco.com); Albert Tian; Eric Rosen; Greg > Shepherd (shep) (shep@cisco.com); Ross Callon > Subject: RE: BIER and deterministic ECMP >=20 > Yes, there is a tradeoff. >=20 > Suppose in your figure that you remove the link from C to Z. In that case= , a > packet destined for Y and Z would be more efficiently sent to D (since on= ly > one copy needs to be sent by A to D, and D can replicate to Y and Z. > However, a packet destined for X and Y would more efficiently be sent to = C, > since A would send only one copy (to C) and C can replicate to X and Y. >=20 > Suppose that X and Z kept joining and leaving the group, so that some > packets go only to X and Y, and some packets go only to Y and Z. In this = case > there is a legitimate question: What behavior do we want? If we minimize > traffic on the links out of A then the packets to X and Y would go via C,= and > the packets to Y and Z would go via D. This would mean that the packets t= o Y > keep fluctuating between paths (or at least fluctuate as quickly as X and= Z > join and leave the group). If we wanted steady performance and > predictability for the traffic to Y, we might prefer to have the traffic = to Y > always take the same path at the cost of sometimes having to send two > packets out of node A. >=20 > Ross >=20 > PS: Your figure showed up and printed fine, at least in my email system. >=20 From nobody Tue Nov 18 10:52:14 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 80C381A6F63 for ; Tue, 18 Nov 2014 10:52:11 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.794 X-Spam-Level: X-Spam-Status: No, score=-4.794 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6gNNzCRe0QA2 for ; Tue, 18 Nov 2014 10:52:02 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 488C01A6F3E for ; Tue, 18 Nov 2014 10:52:02 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml401-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BLT85707; Tue, 18 Nov 2014 18:52:00 +0000 (GMT) Received: from DFWEML706-CHM.china.huawei.com (10.193.5.225) by lhreml401-hub.china.huawei.com (10.201.5.240) with Microsoft SMTP Server (TLS) id 14.3.158.1; Tue, 18 Nov 2014 18:51:59 +0000 Received: from DFWEML703-CHM.china.huawei.com ([10.193.5.130]) by dfweml706-chm ([10.193.5.225]) with mapi id 14.03.0158.001; Tue, 18 Nov 2014 10:51:54 -0800 From: AshwoodsmithPeter To: Antoni Przygienda , "bier@ietf.org" Thread-Topic: Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZAAG90tAACQGRkA= Date: Tue, 18 Nov 2014 18:51:54 +0000 Message-ID: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4D4@dfweml703-chm> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> <2E4BB27CAB87BF43B4207C0E55860F181E75D1@eusaamb103.ericsson.se> In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F181E75D1@eusaamb103.ericsson.se> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.193.60.150] Content-Type: multipart/alternative; boundary="_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4D4dfweml703chm_" MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/0_XhyG1pi7EfMKKfj1f-rrXNhqE Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 18:52:11 -0000 --_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4D4dfweml703chm_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Now does this matter? Well if you are trying to squeeze the last drop of b= andwidth out of the network you need a real Steiner tree for the set of nod= es being connected. Mind you if you are so concerned about bandwidth the 1= 0% or more overhead of the Bier header is likely also a concern. [Tony said] yes, but you don't provide any solution, just raise a problem. = Computing all the Steiners for all sets (which is combinatorial) is not fea= sible obviously. [PETER] Clarifying a problem actually. I raised it in the meeting but was t= old it was not a problem because MT solved it so I am simply clarifying tha= t it is still a problem and MT does not solve it. It may not be a critical = problem but it is none the less true that you could end up with some pretty= horrible trees interconnecting small numbers of nodes. The only solution that I know of along the lines of BIER would be to encode= the links to traverse as a bit set in a header as that would allow an arbi= trary tree per set of destinations. However for that you'd need to have a b= igger header. The alternatives to that are of course the hash based headers= that use Bloom filters but they produce false positives and unwanted traff= ic .. also possible loops. Unfortunately its trade-offs all the way down. I think BIER is actually a p= retty good set of trade-offs but let's be clear on what it can and cannot a= ccomplish. Peter --_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4D4dfweml703chm_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Now does this matter? Well  if you are trying t= o squeeze the last drop of bandwidth out of the network you need a real Ste= iner tree for the set of nodes being connected.  Mind you if you are s= o concerned about bandwidth the 10% or more overhead of the Bier header is likely also a concern.

[Tony said] yes,= but you don’t provide any solution, just raise a problem. Computing = all the Steiners for all sets (which is combinatorial) is not feasible obvi= ously.

 

[PETER] Clarifying a p= roblem actually. I raised it in the meeting but was told it was not a probl= em because MT solved it so I am simply clarifying that it is still a proble= m and MT does not solve it. It may not be a critical problem but it is none the less true that you could end up w= ith some pretty horrible trees interconnecting small numbers of nodes.

 

The only solution that= I know of along the lines of BIER would be to encode the links to traverse= as a bit set in a header as that would allow an arbitrary tree per set of = destinations. However for that you’d need to have a bigger header. The alternatives to that are of course the h= ash based headers that use Bloom filters but they produce false positives a= nd unwanted traffic .. also possible loops.

 

Unfortunately its trad= e-offs all the way down. I think BIER is actually a pretty good set of trad= e-offs but let’s be clear on what it can and cannot accomplish.<= /o:p>

 

Peter

--_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4D4dfweml703chm_-- From nobody Tue Nov 18 10:56:41 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 75ACC1A1B85 for ; Tue, 18 Nov 2014 10:56:37 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.2 X-Spam-Level: X-Spam-Status: No, score=-4.2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id h7MVjdL7O9NR for ; Tue, 18 Nov 2014 10:56:34 -0800 (PST) Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 052961A064C for ; Tue, 18 Nov 2014 10:56:33 -0800 (PST) X-AuditID: c6180641-f79916d00000623a-f3-546b3aeeb303 Received: from EUSAAHC002.ericsson.se (Unknown_Domain [147.117.188.78]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id 17.00.25146.EEA3B645; Tue, 18 Nov 2014 13:26:23 +0100 (CET) Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC002.ericsson.se ([147.117.188.78]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 13:56:32 -0500 From: Antoni Przygienda To: AshwoodsmithPeter , "bier@ietf.org" Thread-Topic: Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZAAG90tAACQGRkAAAFtlEA== Date: Tue, 18 Nov 2014 18:56:31 +0000 Message-ID: <2E4BB27CAB87BF43B4207C0E55860F181E8B39@eusaamb103.ericsson.se> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> <2E4BB27CAB87BF43B4207C0E55860F181E75D1@eusaamb103.ericsson.se> <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4D4@dfweml703-chm> In-Reply-To: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4D4@dfweml703-chm> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [147.117.188.12] Content-Type: multipart/alternative; boundary="_000_2E4BB27CAB87BF43B4207C0E55860F181E8B39eusaamb103ericsso_" MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrNLMWRmVeSWpSXmKPExsUyuXSPn+57q+wQg9dHWSyWztjDZLHwbhuT A5NHy5G3rB5LlvxkCmCK4rJJSc3JLEst0rdL4Mr4vWERW0FHZMX7o1NZGhhP+XYxcnJICJhI HJl8gx3CFpO4cG89WxcjF4eQwBFGiS8rXjJBOMsZJRrbnjKCVLEJWEhc/vaUGcQWEYiQ6LnT DBTn4BAWyJT4/MMPxBQRyJJ4+NENosJJomVyHxOIzSKgKnH491qwKbwC3hI7DuwA2yskcJBR Yvs8IxCbU8BVYvnHO2DTGYHu+X5qDVgvs4C4xK0n85kg7hSQWLLnPDOELSrx8vE/VghbSWLO 62vMEPX5Ej/a37ND7BKUODnzCcsERpFZSEbNQlI2C0kZRFxHYsHuT2wQtrbEsoWvmWHsMwce MyGLL2BkX8XIUVqcWpabbmS4iREYO8ck2Bx3MC74ZHmIUYCDUYmHd4NrVogQa2JZcWXuIUZp DhYlcV7N6nnBQgLpiSWp2ampBalF8UWlOanFhxiZODilGhgPb000E5DN/7LPfqq0oanJ/8SW GPcJdmwZ6UWX/IOU83xSHnA7C14vuzpnIkeOlDPTrBcOSXdjXt/fOOnlxiLd8quV+/svLajX vf9oS4BnTW+dZX2WyL30/mPpdmXJZdcsDye+23uZd5m47/YJN92PzJnYdyEjN+PQef6PqkW+ bi1FtUZ/o5VYijMSDbWYi4oTAXvI4/5+AgAA Archived-At: http://mailarchive.ietf.org/arch/msg/bier/X9XkMMJyD48F2z88UMo75dIyxX8 Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 18:56:37 -0000 --_000_2E4BB27CAB87BF43B4207C0E55860F181E8B39eusaamb103ericsso_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Peter, yes, agreed, very fair summary. I didn't say in the meeting that M= T solves it, must have been someone else ;-) I only clarified with you that I considered the 'Steiner problem' a single= Steiner for _all_ I/E on the tree so we talked about different assumptions= . I actually _can_ imagine a variation of BIER encaps that would list the lin= ks but I think that would be truly 'multicast segment routing' rather than= BIER as was proposed in Hawaii ;-) tony From: AshwoodsmithPeter [mailto:Peter.AshwoodSmith@huawei.com] Sent: Tuesday, November 18, 2014 10:52 AM To: Antoni Przygienda; bier@ietf.org Subject: RE: Follow up on my comments from Honolulu concerning T.E/ Steiner= Tree optimization with Bier Now does this matter? Well if you are trying to squeeze the last drop of b= andwidth out of the network you need a real Steiner tree for the set of nod= es being connected. Mind you if you are so concerned about bandwidth the 1= 0% or more overhead of the Bier header is likely also a concern. [Tony said] yes, but you don't provide any solution, just raise a problem. = Computing all the Steiners for all sets (which is combinatorial) is not fea= sible obviously. [PETER] Clarifying a problem actually. I raised it in the meeting but was t= old it was not a problem because MT solved it so I am simply clarifying tha= t it is still a problem and MT does not solve it. It may not be a critical = problem but it is none the less true that you could end up with some pretty= horrible trees interconnecting small numbers of nodes. The only solution that I know of along the lines of BIER would be to encode= the links to traverse as a bit set in a header as that would allow an arbi= trary tree per set of destinations. However for that you'd need to have a b= igger header. The alternatives to that are of course the hash based headers= that use Bloom filters but they produce false positives and unwanted traff= ic .. also possible loops. Unfortunately its trade-offs all the way down. I think BIER is actually a p= retty good set of trade-offs but let's be clear on what it can and cannot a= ccomplish. Peter --_000_2E4BB27CAB87BF43B4207C0E55860F181E8B39eusaamb103ericsso_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

= Peter, yes, agreed, very fai= r summary.   I didn’t say in the meeting that MT solves it,= must have been someone else ;-)  

= I only clarified with you th= at  I considered the ‘Steiner problem’ a single Steiner fo= r _all_ I/E on the tree so we talked about different assumptions.

=  

= I actually _can_ imagine a variation of BIER encaps t= hat would list the links but I think that would be truly  ‘multicast segment routing’ rather than BIER

= as was proposed in Hawaii ;-= )

=  

= tony

=  

From: AshwoodsmithPeter [mailto:Peter.AshwoodSmith@huawei.com]
Sent: Tuesday, November 18, = 2014 10:52 AM
To: Antoni Przygienda; bier@= ietf.org
Subject: RE: Follow up on my= comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier=

 

Now does this matter? Well  if you are trying to squeez= e the last drop of bandwidth out of the network you need a real Steiner tre= e for the set of nodes being connected.  Mind you if you are so concerned about bandwidth the 10% or more overhead of th= e Bier header is likely also a concern.

[Tony said] yes, but you don’t provide any solution, jus= t raise a problem. Computing all the Steiners for all sets (which is combinatorial) is not feasible obviously. <= /span>

=  

= [PETER] Clarifying a problem= actually. I raised it in the meeting but was told it was not a problem bec= ause MT solved it so I am simply clarifying that it is still a problem and MT does not solve it. It may not be a criti= cal problem but it is none the less true that you could end up with some pr= etty horrible trees interconnecting small numbers of nodes.

=  

= The only solution that I kno= w of along the lines of BIER would be to encode the links to traverse as a = bit set in a header as that would allow an arbitrary tree per set of destinations. However for that you’d need = to have a bigger header. The alternatives to that are of course the hash ba= sed headers that use Bloom filters but they produce false positives and unw= anted traffic .. also possible loops.

=  

= Unfortunately its trade-offs= all the way down. I think BIER is actually a pretty good set of trade-offs= but let’s be clear on what it can and cannot accomplish.

=  

= Peter

--_000_2E4BB27CAB87BF43B4207C0E55860F181E8B39eusaamb103ericsso_-- From nobody Tue Nov 18 11:00:34 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3BFC81A6F7D for ; Tue, 18 Nov 2014 11:00:31 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.795 X-Spam-Level: X-Spam-Status: No, score=-4.795 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id VCqQFnwmzfQH for ; Tue, 18 Nov 2014 11:00:28 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 6FB8B1A6F8C for ; Tue, 18 Nov 2014 11:00:27 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BOY15785; Tue, 18 Nov 2014 19:00:25 +0000 (GMT) Received: from DFWEML706-CHM.china.huawei.com (10.193.5.225) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Tue, 18 Nov 2014 19:00:24 +0000 Received: from DFWEML703-CHM.china.huawei.com ([10.193.5.130]) by dfweml706-chm ([10.193.5.225]) with mapi id 14.03.0158.001; Tue, 18 Nov 2014 11:00:14 -0800 From: AshwoodsmithPeter To: "bier@ietf.org" Thread-Topic: [Bier] BIER and deterministic ECMP Thread-Index: AQHQAvPmimRcG0ie5kmRBDyo9YL80ZxmvLFw Date: Tue, 18 Nov 2014 19:00:14 +0000 Message-ID: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4EE@dfweml703-chm> References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.193.60.150] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/s-Lmdkpl5sCvm4T3-YQ5cgDWpek Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 19:00:31 -0000 What if rather than process in the order of the bits the processing is inve= rted and instead the links are processed in some lexicographic ordering as = the outer loop. That would lead to a determinism of the paths used irrespec= tive of entry or departure of lower order bits. Slower I guess for a sparse= ly populated bit set. Peter=20 -----Original Message----- From: BIER [mailto:bier-bounces@ietf.org] On Behalf Of Albert Tian Sent: Tuesday, November 18, 2014 12:53 AM To: Ross Callon; bier@ietf.org Subject: Re: [Bier] BIER and deterministic ECMP Hi Ross, Very good point. Yes IMHO this needs to be addressed, otherwise the links l= eading to lower numbered BFERs will more likely be chosen in ECMP and cause= unbalanced traffic. This maybe just as bad as the instability you have ide= ntified. Agreed multiple BIFT solution should work. Thanks, Albert On 11/17/14, 3:10 PM, "Ross Callon" wrote: >Reading the BIER architecture (draft-wijnands-bier-architecture-01) I=20 >noticed the last sentence of section 6.7: > > Note however that by the rules of Section 6.5, any packet destined > for both BFER-D and BFER-F will be sent via BFR-C. > >Thinking about this more: It is clear that where ECMP paths exist,=20 >which path is taken to reach any particular destination will depend=20 >upon which other destinations the same packet goes to. In particular,=20 >which lower numbered bits are set in the BitString can change the=20 >choice among ECMP paths. This could have the effect that as a lower=20 >numbered destination intermittently joins and leaves a multicast group,=20 >the path to a higher numbered destination could switch back and forth=20 >between different ECMP paths. >=20 >This could be fixed if we used multiple BIFT tables, such that each=20 >BIFT table has only one path to each destination, but the multiple=20 >possible ECMP paths to any particular destination are spread across the=20 >multiple tables. When a BIER-encapsulated packet arrives to be=20 >forwarded, the BFR would use the BIER Entropy field (or a hash thereon,=20 >if fewer than 256 next hops are >supported) >to determine which BIFT to use, and then the normal BIER forwarding=20 >algorithm (the same as described in=20 >draft-wijnands-bier-architecture-01) would be used with whichever BIFT=20 >table was selected. This would of course increase the amount of memory=20 >that would be needed to store BIFT tables, but my quick "back of the=20 >envelope" calculation suggests that the amount of memory to store BIFT=20 >tables still should not be a problem if there aren't too many BIFT=20 >tables (I did my "back of the envelope" with max 16 tables). The=20 >forwarding algorithm and packet headers stay unchanged. This also has=20 >the feature that if this approach is consistently implemented across a=20 >service provider, then where a problem has been reported on a=20 >particular flow of packets to a particular receiver, we can traceroute=20 >with the same entropy value to that particular receiver and know that=20 >we will be using the same path as the normal data packets to that=20 >receiver. >=20 >I would be happy to write up appropriate text to add to >draft-wijnands-bier- >architecture-01 (presumably for a -02 version) if the authors would=20 >like, or alternately could write up a separate Internet Draft to=20 >further describe the issue and the solution. > >Thanks, Ross > > > >_______________________________________________ >BIER mailing list >BIER@ietf.org >https://www.ietf.org/mailman/listinfo/bier _______________________________________________ BIER mailing list BIER@ietf.org https://www.ietf.org/mailman/listinfo/bier From nobody Tue Nov 18 11:14:34 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 84A271A1B85 for ; Tue, 18 Nov 2014 11:14:31 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.902 X-Spam-Level: X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id LHcFibKgaaRE for ; Tue, 18 Nov 2014 11:14:26 -0800 (PST) Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0144.outbound.protection.outlook.com [65.55.169.144]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 93A781A1B67 for ; Tue, 18 Nov 2014 11:14:24 -0800 (PST) Received: from BY1PR0501MB1430.namprd05.prod.outlook.com (25.160.107.152) by BY1PR0501MB1431.namprd05.prod.outlook.com (25.160.107.153) with Microsoft SMTP Server (TLS) id 15.1.16.15; Tue, 18 Nov 2014 19:14:22 +0000 Received: from BY1PR0501MB1430.namprd05.prod.outlook.com ([25.160.107.152]) by BY1PR0501MB1430.namprd05.prod.outlook.com ([25.160.107.152]) with mapi id 15.01.0016.006; Tue, 18 Nov 2014 19:14:22 +0000 From: Ross Callon To: AshwoodsmithPeter Thread-Topic: [Bier] BIER and deterministic ECMP Thread-Index: AQHQAvPmimRcG0ie5kmRBDyo9YL80ZxmvLFwgAAB7QA= Date: Tue, 18 Nov 2014 19:14:22 +0000 Message-ID: References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4EE@dfweml703-chm> In-Reply-To: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4EE@dfweml703-chm> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [66.129.241.13] x-microsoft-antispam: BCL:0;PCL:0;RULEID:;SRVR:BY1PR0501MB1431; x-exchange-antispam-report-test: UriScan:; x-exchange-antispam-report-cfa-test: BCL:0; PCL:0; RULEID:; SRVR:BY1PR0501MB1431; x-forefront-prvs: 039975700A x-forefront-antispam-report: SFV:NSPM; SFS:(10019020)(6009001)(164054003)(51704005)(24454002)(199003)(13464003)(189002)(377454003)(479174003)(106356001)(76176999)(50986999)(66066001)(107046002)(2656002)(105586002)(99286002)(87936001)(64706001)(76576001)(20776003)(21056001)(54356999)(92566001)(106116001)(86362001)(74316001)(110136001)(15975445006)(19580405001)(19580395003)(101416001)(122556002)(33646002)(4396001)(95666004)(77096003)(62966003)(108616004)(97736003)(99396003)(46102003)(40100003)(31966008)(120916001)(77156002)(24736002); DIR:OUT; SFP:1102; SCL:1; SRVR:BY1PR0501MB1431; H:BY1PR0501MB1430.namprd05.prod.outlook.com; FPR:; MLV:sfv; PTR:InfoNoRecords; A:1; MX:1; LANG:en; Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-OriginatorOrg: juniper.net Archived-At: http://mailarchive.ietf.org/arch/msg/bier/IfoRrs6vow7fnLu3on7S3o1rBXI Cc: Ross Callon , "bier@ietf.org" Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 19:14:32 -0000 I think that the problem with this is that the number of links out of the r= outer can be quite large. You may be forwarding any one packet on only a sm= all number of links, and want to be able to keep all of your outgoing links= full. Therefore you want to have a reasonable limit on the number of looku= ps required for each packet. A good feature of the algorithm currently desc= ribed in the BIER architecture is that the number of lookups that you need = to do is bounded by the number of copies of the packet that you will actual= ly be sending out.=20 Also, in the example that Tony and I were discussing, regardless of what lo= okup algorithm is used, you still have the issue of whether you want the pa= ckets to Y to all go out the same way (deterministic forwarding), or do you= want the path to fluctuate based on what other destinations are receiving = the packet (minimize the number of packets transmitted).=20 Ross -----Original Message----- From: BIER [mailto:bier-bounces@ietf.org] On Behalf Of AshwoodsmithPeter Sent: Tuesday, November 18, 2014 2:00 PM To: bier@ietf.org Subject: Re: [Bier] BIER and deterministic ECMP What if rather than process in the order of the bits the processing is inve= rted and instead the links are processed in some lexicographic ordering as = the outer loop. That would lead to a determinism of the paths used irrespec= tive of entry or departure of lower order bits. Slower I guess for a sparse= ly populated bit set. Peter=20 -----Original Message----- From: BIER [mailto:bier-bounces@ietf.org] On Behalf Of Albert Tian Sent: Tuesday, November 18, 2014 12:53 AM To: Ross Callon; bier@ietf.org Subject: Re: [Bier] BIER and deterministic ECMP Hi Ross, Very good point. Yes IMHO this needs to be addressed, otherwise the links l= eading to lower numbered BFERs will more likely be chosen in ECMP and cause= unbalanced traffic. This maybe just as bad as the instability you have ide= ntified. Agreed multiple BIFT solution should work. Thanks, Albert On 11/17/14, 3:10 PM, "Ross Callon" wrote: >Reading the BIER architecture (draft-wijnands-bier-architecture-01) I=20 >noticed the last sentence of section 6.7: > > Note however that by the rules of Section 6.5, any packet destined > for both BFER-D and BFER-F will be sent via BFR-C. > >Thinking about this more: It is clear that where ECMP paths exist,=20 >which path is taken to reach any particular destination will depend=20 >upon which other destinations the same packet goes to. In particular,=20 >which lower numbered bits are set in the BitString can change the=20 >choice among ECMP paths. This could have the effect that as a lower=20 >numbered destination intermittently joins and leaves a multicast group,=20 >the path to a higher numbered destination could switch back and forth=20 >between different ECMP paths. >=20 >This could be fixed if we used multiple BIFT tables, such that each=20 >BIFT table has only one path to each destination, but the multiple=20 >possible ECMP paths to any particular destination are spread across the=20 >multiple tables. When a BIER-encapsulated packet arrives to be=20 >forwarded, the BFR would use the BIER Entropy field (or a hash thereon,=20 >if fewer than 256 next hops are >supported) >to determine which BIFT to use, and then the normal BIER forwarding=20 >algorithm (the same as described in=20 >draft-wijnands-bier-architecture-01) would be used with whichever BIFT=20 >table was selected. This would of course increase the amount of memory=20 >that would be needed to store BIFT tables, but my quick "back of the=20 >envelope" calculation suggests that the amount of memory to store BIFT=20 >tables still should not be a problem if there aren't too many BIFT=20 >tables (I did my "back of the envelope" with max 16 tables). The=20 >forwarding algorithm and packet headers stay unchanged. This also has=20 >the feature that if this approach is consistently implemented across a=20 >service provider, then where a problem has been reported on a=20 >particular flow of packets to a particular receiver, we can traceroute=20 >with the same entropy value to that particular receiver and know that=20 >we will be using the same path as the normal data packets to that=20 >receiver. >=20 >I would be happy to write up appropriate text to add to >draft-wijnands-bier- >architecture-01 (presumably for a -02 version) if the authors would=20 >like, or alternately could write up a separate Internet Draft to=20 >further describe the issue and the solution. > >Thanks, Ross > > > >_______________________________________________ >BIER mailing list >BIER@ietf.org >https://www.ietf.org/mailman/listinfo/bier _______________________________________________ BIER mailing list BIER@ietf.org https://www.ietf.org/mailman/listinfo/bier _______________________________________________ BIER mailing list BIER@ietf.org https://www.ietf.org/mailman/listinfo/bier From nobody Tue Nov 18 11:24:49 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8A67D1A1B67 for ; Tue, 18 Nov 2014 11:24:37 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.201 X-Spam-Level: X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Owhw9x6hkVjL for ; Tue, 18 Nov 2014 11:24:34 -0800 (PST) Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B51B11A6F3D for ; Tue, 18 Nov 2014 11:24:13 -0800 (PST) X-AuditID: c618062d-f79206d0000014d2-7c-546b43c65186 Received: from EUSAAHC004.ericsson.se (Unknown_Domain [147.117.188.84]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id A3.68.05330.6C34B645; Tue, 18 Nov 2014 14:04:06 +0100 (CET) Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC004.ericsson.se ([147.117.188.84]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 14:23:58 -0500 From: Antoni Przygienda To: Ross Callon , AshwoodsmithPeter Thread-Topic: [Bier] BIER and deterministic ECMP Thread-Index: AQHQAvPmimRcG0ie5kmRBDyo9YL80ZxmvLFwgAAB7QCAAATaQA== Date: Tue, 18 Nov 2014 19:23:57 +0000 Message-ID: <2E4BB27CAB87BF43B4207C0E55860F181E8B89@eusaamb103.ericsson.se> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4EE@dfweml703-chm> In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [147.117.188.12] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFnrOLMWRmVeSWpSXmKPExsUyuXRPiO4x5+wQg/NNNhZLZ+xhslh4t43J 4u+KKywOzB4tR96yeixZ8pPJ43rTVfYA5igum5TUnMyy1CJ9uwSujBkrFAru8lTcmlnfwNjP 1cXIySEhYCKx6G4HM4QtJnHh3nq2LkYuDiGBI4wS05seMEE4yxklZl/azQhSxSZgIXH521Ow DhGBKInpPReYQGxmAWWJz5sngdnCAvoSF5Z+ZOli5ACqMZDobuKCKHeS+Db3KztImEVAVaJp ZS2IySvgLTHtqCzEpgOMEquf/mMFKecUiJQ42H+UDcRmBLrt+6k1UJvEJW49mc8EcbOAxJI9 56HuF5V4+RiiV0JASWLO62vMEPU6Egt2f2KDsLUlli18DRbnFRCUODnzCcsERrFZSMbOQtIy C0nLLCQtCxhZVjFylBanluWmGxlsYgTGzDEJNt0djHteWh5iFOBgVOLh3eCaFSLEmlhWXJl7 iFGag0VJnHdW7bxgIYH0xJLU7NTUgtSi+KLSnNTiQ4xMHJxSDYyOTnWLL3nHPZ9tsXVf7fHp 34R3vkkr06/4s7ihhkn+Szb/ATOVByWMKTGRFq7h9l1Sbwwztj9/wyKrzTNNkt/wkfO/Sfqr g41/L1j4vJuLu/7gzxa2h694Lu+unGS3ItHFxMz3zwIVE87D6lzPtfw4rkucOv109bZH1udV JWbEO1stMpWzvqrEUpyRaKjFXFScCADVcwRpegIAAA== Archived-At: http://mailarchive.ietf.org/arch/msg/bier/P4sK8jtcfNkRhk4ozi1M58JaQAw Cc: "bier@ietf.org" Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 19:24:37 -0000 Yes, there are all kind of 'dialects' of the algorithm that can be imagined= , e.g.=20 matching the 'entry with most bits' that AND's with the packet mask first i= nstead of 'right side' walking yields quite interesting properties when deployed = network-wide. If one can hire a hardware guy smart and cheap=20 enough, one could build a TCAM variation just for that ;-)=20 --- tony =20 > -----Original Message----- > From: BIER [mailto:bier-bounces@ietf.org] On Behalf Of Ross Callon > Sent: Tuesday, November 18, 2014 11:14 AM > To: AshwoodsmithPeter > Cc: Ross Callon; bier@ietf.org > Subject: Re: [Bier] BIER and deterministic ECMP >=20 > I think that the problem with this is that the number of links out of the > router can be quite large. You may be forwarding any one packet on only a > small number of links, and want to be able to keep all of your outgoing l= inks > full. Therefore you want to have a reasonable limit on the number of > lookups required for each packet. A good feature of the algorithm current= ly > described in the BIER architecture is that the number of lookups that you > need to do is bounded by the number of copies of the packet that you will > actually be sending out. >=20 > Also, in the example that Tony and I were discussing, regardless of what > lookup algorithm is used, you still have the issue of whether you want th= e > packets to Y to all go out the same way (deterministic forwarding), or do= you > want the path to fluctuate based on what other destinations are receiving > the packet (minimize the number of packets transmitted). >=20 From nobody Tue Nov 18 11:28:01 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B696E1A6FCD for ; Tue, 18 Nov 2014 11:27:59 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.795 X-Spam-Level: X-Spam-Status: No, score=-4.795 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XSwCf26kXT1t for ; Tue, 18 Nov 2014 11:27:57 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id EE2201A6FC9 for ; Tue, 18 Nov 2014 11:27:35 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml403-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BOY17173; Tue, 18 Nov 2014 19:27:34 +0000 (GMT) Received: from DFWEML706-CHM.china.huawei.com (10.193.5.225) by lhreml403-hub.china.huawei.com (10.201.5.217) with Microsoft SMTP Server (TLS) id 14.3.158.1; Tue, 18 Nov 2014 19:27:33 +0000 Received: from DFWEML703-CHM.china.huawei.com ([10.193.5.130]) by dfweml706-chm ([10.193.5.225]) with mapi id 14.03.0158.001; Tue, 18 Nov 2014 11:27:23 -0800 From: AshwoodsmithPeter To: Eric C Rosen , "bier@ietf.org" Thread-Topic: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZAA1gi8AAAmjE1A= Date: Tue, 18 Nov 2014 19:27:22 +0000 Message-ID: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D531@dfweml703-chm> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> <546B6938.9020709@juniper.net> In-Reply-To: <546B6938.9020709@juniper.net> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.193.60.150] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/c7-8EBDT353KfJoWSiK5VsROr9o Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 19:27:59 -0000 Inline [PETER] > -----Original Message----- > From: Eric C Rosen [mailto:erosen@juniper.net] > Sent: Tuesday, November 18, 2014 10:44 AM > To: AshwoodsmithPeter; bier@ietf.org > Cc: erosen@juniper.net > Subject: Re: [Bier] Follow up on my comments from Honolulu concerning > T.E/ Steiner Tree optimization with Bier >=20 > > To compute the Steiner tree you therefore need the exact set of nodes > > to be connected as input. Since Bier routing has no knowledge of the > > set of nodes being connected when it computes the path to a given > node > > for a given MT it is impossible for those computations to produce a > > proper Steiner tree for each subset of nodes. >=20 > This is an interesting point, though I don't think it's stated quite > right. A phrase like "BIER routing has no knowledge of the set of > nodes being connected when it computes the path" is a bit imprecise, > given the three-layer BIER architecture. [PETER] Ok, more precisely the computation of the next hops has no knowledg= e of all the intended member subsets. > In at least some applications, the multicast flow layer will know the > exact set of nodes that needs to be connected for a given multicast > flow. It is then possible for the multicast flow layer to cause a > Steiner tree for that flow to be created. That tree can then be > identified (e.g., assigned an MT-id), and used as an underlay by the > BIER forwarding procedures. So I don't think there is anything in the > BIER architecture that prevents the use of Steiner trees. [PETER] True but that's a bit of a stretch. In that case the MT computed tr= ee creates state for the use of only that set of destinations and is essent= ially removing the stateless (per set of destinations) nature of Bier. > However ... >=20 > If you're going to build a tree that contains all and only the nodes > through which a given flow needs to travel, you can use conventional > multicast forwarding to send packets of that flow on that tree. BIER > doesn't provide any added value in that case. [PETER] Correct, my point above. Works but is not really saving you any sta= te and on the contrary the 10% of so B/W tax is still present so it not onl= y does not buy you anything it costs you something that other trees don't. > Alternatively, as Tony suggests, one could build a Steiner tree > connecting all the BFIRs/BFERs in the domain. Then one could send all > flows along that tree, using BIER to prevent a given flow from going > through nodes that it does not need to visit. I suspect Peter would > say though that there is no point in doing this, as the path taken by > any given flow would not be optimized (i.e., the path traveled by a > given flow is not the path along the Steiner tree for that flow, and > hence does not have the properties one might hope for). [PETER] Would say there is likely a very good point in doing that much of t= he time but there will be some horrific degenerate cases with small numbers= of members out in the edges of that tree that will surprise people.=20 Very likely what you want is to have a number of MT's where each MT has a d= ifferent 'center of gravity' for want of a better word. When dealing with s= mall sets of members that are localized, the centers of gravity of that set= of members is used to pick an appropriate MT, thus minimizing the horrific= kinds of routing that would otherwise result. So a bit of offline analysis= will be required to pick a good MT and MT's will need to be distributed pr= operly to avoid bad degenerate routing. > So I think the architecture does support the use case that Tony > describes; whether that use case is actually useful or not is something > we don't have to decide. [PETER] With sufficient MT's even Bier will fly ;)=20 Sorry could not resist :) Again I'm not knocking the idea, think it's pretty cool and hope it goes ah= ead, just trying to be clear on what it can and cannot do and how it would = have to be used to get results similar to say P2MP central control. Peter From nobody Tue Nov 18 11:31:27 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1BE131A6FB6 for ; Tue, 18 Nov 2014 11:31:25 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.201 X-Spam-Level: X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qXYrtv_YFoS1 for ; Tue, 18 Nov 2014 11:31:23 -0800 (PST) Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C6ACB1A6F8F for ; Tue, 18 Nov 2014 11:31:22 -0800 (PST) X-AuditID: c618062d-f79206d0000014d2-62-546b457ee832 Received: from EUSAAHC005.ericsson.se (Unknown_Domain [147.117.188.87]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id 55.D8.05330.E754B645; Tue, 18 Nov 2014 14:11:27 +0100 (CET) Received: from EUSAAMB103.ericsson.se ([147.117.188.120]) by EUSAAHC005.ericsson.se ([147.117.188.87]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 14:31:21 -0500 From: Antoni Przygienda To: AshwoodsmithPeter , Eric C Rosen , "bier@ietf.org" Thread-Topic: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZAAvONsAAAfOPwAACmadAA== Date: Tue, 18 Nov 2014 19:31:20 +0000 Message-ID: <2E4BB27CAB87BF43B4207C0E55860F181E8BA5@eusaamb103.ericsson.se> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> <546B6938.9020709@juniper.net> <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D531@dfweml703-chm> In-Reply-To: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D531@dfweml703-chm> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [147.117.188.12] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFrrHLMWRmVeSWpSXmKPExsUyuXRPuG69a3aIwb9mRoulM/YwWazb8IHZ YuHdNiYHZo+WI29ZPZYs+cnkcb3pKnsAcxSXTUpqTmZZapG+XQJXxox7E5kLNrNUzNp+k7WB cRtzFyMnh4SAicSj7VtZIWwxiQv31rN1MXJxCAkcYZQ49/EPO4SznFHi5MSZLCBVbAIWEpe/ PQXrFhGolniy/SpQNweHsECBxJL9KhDhQonvOz5DlThJ7Dk7F2wBi4CqxIOTL9hBbF4Bb4kV P88zQ8xfwCjRNauPDWQOp4CrxPeX2SA1jEAHfT+1hgnEZhYQl7j1ZD4TxKECEkv2nId6QFTi 5eN/UA8oScx5fY0Zol5HYsHuT2wQtrbEsoWvmSH2CkqcnPmEZQKj6CwkY2chaZmFpGUWkpYF jCyrGDlKi1PLctONDDYxAiPkmASb7g7GPS8tDzEKcDAq8fBucM0KEWJNLCuuzD3EKM3BoiTO O6t2XrCQQHpiSWp2ampBalF8UWlOavEhRiYOTqkGRunc/1E8XbfW8tx8Fp1W9FXmxBOLLFXu gLQnidP496cV9qbcsODdOUtAjb2dO+bbqvQCbuV3/foaLAs9hVis9JKTnkbdNnCMmsKlc63s CYPb2rcu387PZJ+w7dVfj6hdjlz8iw/N+8WUI+IhmaItuG/nvm0TV05KbpaOirW6fFf425I3 M5x3KrEUZyQaajEXFScCABc9EnRxAgAA Archived-At: http://mailarchive.ietf.org/arch/msg/bier/DS_aMXG5JEJesZ8zAyf95JvM6pw Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 19:31:25 -0000 >=20 > Very likely what you want is to have a number of MT's where each MT has a > different 'center of gravity' for want of a better word. When dealing wit= h > small sets of members that are localized, the centers of gravity of that = set of > members is used to pick an appropriate MT, thus minimizing the horrific > kinds of routing that would otherwise result. So a bit of offline analysi= s will > be required to pick a good MT and MT's will need to be distributed proper= ly > to avoid bad degenerate routing. >=20 [Tony said] RFC2201 ;-) From nobody Tue Nov 18 11:38:33 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 675321A7012 for ; Tue, 18 Nov 2014 11:38:29 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.902 X-Spam-Level: X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fQ8ic7tHuwyh for ; Tue, 18 Nov 2014 11:38:27 -0800 (PST) Received: from na01-bl2-obe.outbound.protection.outlook.com (mail-bl2on0109.outbound.protection.outlook.com [65.55.169.109]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id B5E851A7022 for ; Tue, 18 Nov 2014 11:38:26 -0800 (PST) Received: from CY1PR0501MB1097.namprd05.prod.outlook.com (25.160.144.139) by CY1PR0501MB1195.namprd05.prod.outlook.com (25.160.145.144) with Microsoft SMTP Server (TLS) id 15.1.16.15; Tue, 18 Nov 2014 19:38:24 +0000 Received: from [172.29.32.63] (66.129.241.11) by CY1PR0501MB1097.namprd05.prod.outlook.com (25.160.144.139) with Microsoft SMTP Server (TLS) id 15.1.16.15; Tue, 18 Nov 2014 19:38:23 +0000 Message-ID: <546BA029.8070208@juniper.net> Date: Tue, 18 Nov 2014 14:38:17 -0500 From: Eric C Rosen User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.2.0 MIME-Version: 1.0 To: Antoni Przygienda , Ross Callon , "bier@ietf.org" References: <2E4BB27CAB87BF43B4207C0E55860F181E8A95@eusaamb103.ericsson.se> <3f8182b06bf741e3ad3e750db7ded76a@BY1PR0501MB1430.namprd05.prod.outlook.com> <2E4BB27CAB87BF43B4207C0E55860F181E8B1A@eusaamb103.ericsson.se> In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F181E8B1A@eusaamb103.ericsson.se> Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: [66.129.241.11] X-ClientProxiedBy: DM2PR11CA0026.namprd11.prod.outlook.com (25.160.91.36) To CY1PR0501MB1097.namprd05.prod.outlook.com (25.160.144.139) X-Microsoft-Antispam: UriScan:;UriScan:; X-Microsoft-Antispam: BCL:0;PCL:0;RULEID:;SRVR:CY1PR0501MB1097; X-Exchange-Antispam-Report-Test: UriScan:; X-Exchange-Antispam-Report-CFA-Test: BCL:0; PCL:0; RULEID:; SRVR:CY1PR0501MB1097; X-Forefront-PRVS: 039975700A X-Forefront-Antispam-Report: SFV:NSPM; SFS:(10019020)(6049001)(6009001)(24454002)(189002)(199003)(479174003)(377454003)(86362001)(92566001)(102836001)(93886004)(50986999)(50466002)(40100003)(77156002)(62966003)(77096003)(64126003)(83506001)(80316001)(2501002)(92726001)(59896002)(47776003)(66066001)(46102003)(33656002)(65816999)(20776003)(87266999)(23746002)(54356999)(65806001)(122386002)(76176999)(64706001)(120916001)(4396001)(101416001)(99396003)(87976001)(65956001)(31966008)(95666004)(107046002)(106356001)(21056001)(42186005)(105586002)(97736003)(36756003); DIR:OUT; SFP:1102; SCL:1; SRVR:CY1PR0501MB1097; H:[172.29.32.63]; FPR:; MLV:sfv; PTR:InfoNoRecords; MX:1; A:1; LANG:en; X-Exchange-Antispam-Report-CFA-Test: BCL:0; PCL:0; RULEID:; SRVR:CY1PR0501MB1097; X-Microsoft-Antispam: BCL:0;PCL:0;RULEID:;SRVR:CY1PR0501MB1195; X-OriginatorOrg: juniper.net Archived-At: http://mailarchive.ietf.org/arch/msg/bier/BIInl_o9YRTsRvkDPLIVxhLfx8M Cc: Albert Tian , "Greg Shepherd \(shep\) \(shep@cisco.com\)" , "IJsbrand Wijnands \(ice@cisco.com\)" , erosen@juniper.net Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 19:38:29 -0000 On 11/18/2014 1:50 PM, Antoni Przygienda wrote: > The only property I see as uniformly desirable is that while the > receiver set and label stack is stable, OAM must be deterministic as > to which of the ECMP paths it's following I don't agree. I think the path for a flow to travel from A to B needs to be independent of whether or not there are other receivers. Suppose in your diagram that traffic from A to Y along takes the path A-->D-->Y. But traffic from A to (X+Y) goes from A to C and then from C to X and from C to Y. Now suppose that the link from C to Y is flaky. This means that whenever someone at site X joins the flow, the folks at site Y will see a degradation in service. When the folks at site X stop receiving the flow, the folks at site Y will see an improvement in service. This seems like something that would be very hard to troubleshoot, and could really annoy the folks at site Y. > I think it's pretty much a vendor optimization we're talking here I think it's more a question of what sort of behavior the operators want. From nobody Tue Nov 18 11:47:28 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9380D1A86E9 for ; Tue, 18 Nov 2014 11:47:26 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.794 X-Spam-Level: X-Spam-Status: No, score=-4.794 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WjwAuxwogD1C for ; Tue, 18 Nov 2014 11:47:23 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 64EC71A6FB6 for ; Tue, 18 Nov 2014 11:47:22 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml404-hub.china.huawei.com) ([172.18.7.190]) by lhrrg02-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BLT88501; Tue, 18 Nov 2014 19:47:20 +0000 (GMT) Received: from DFWEML704-CHM.china.huawei.com (10.193.5.141) by lhreml404-hub.china.huawei.com (10.201.5.218) with Microsoft SMTP Server (TLS) id 14.3.158.1; Tue, 18 Nov 2014 19:47:20 +0000 Received: from DFWEML703-CHM.china.huawei.com ([10.193.5.130]) by dfweml704-chm ([10.193.5.141]) with mapi id 14.03.0158.001; Tue, 18 Nov 2014 11:47:06 -0800 From: AshwoodsmithPeter To: Antoni Przygienda , "bier@ietf.org" Thread-Topic: Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZAAG90tAACQGRkAAAFtlEAABv4cg Date: Tue, 18 Nov 2014 19:47:06 +0000 Message-ID: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D5A2@dfweml703-chm> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> <2E4BB27CAB87BF43B4207C0E55860F181E75D1@eusaamb103.ericsson.se> <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4D4@dfweml703-chm> <2E4BB27CAB87BF43B4207C0E55860F181E8B39@eusaamb103.ericsson.se> In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F181E8B39@eusaamb103.ericsson.se> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.193.60.150] Content-Type: multipart/alternative; boundary="_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D5A2dfweml703chm_" MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/Z5KeA9Hb_89VDo0dM3gMqCl1W8E Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 19:47:26 -0000 --_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D5A2dfweml703chm_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Exactly, if you assign a bit per adjacency and the header encodes the set o= f adjacencies to cross then you get segment routing for multicast. I.e. sou= rce routed multicast and can create any tree you want ingress. Pretty cool = if you can afford the overhead. I suppose you could compress with the heade= r encoding logical adjacencies into some set of abstract topologies .. prob= ably give you similar behavior to the MT solution. Peter From: Antoni Przygienda [mailto:antoni.przygienda@ericsson.com] Sent: Tuesday, November 18, 2014 1:57 PM To: AshwoodsmithPeter; bier@ietf.org Subject: RE: Follow up on my comments from Honolulu concerning T.E/ Steiner= Tree optimization with Bier Peter, yes, agreed, very fair summary. I didn't say in the meeting that M= T solves it, must have been someone else ;-) I only clarified with you that I considered the 'Steiner problem' a single= Steiner for _all_ I/E on the tree so we talked about different assumptions= . I actually _can_ imagine a variation of BIER encaps that would list the lin= ks but I think that would be truly 'multicast segment routing' rather than= BIER as was proposed in Hawaii ;-) tony From: AshwoodsmithPeter [mailto:Peter.AshwoodSmith@huawei.com] Sent: Tuesday, November 18, 2014 10:52 AM To: Antoni Przygienda; bier@ietf.org Subject: RE: Follow up on my comments from Honolulu concerning T.E/ Steiner= Tree optimization with Bier Now does this matter? Well if you are trying to squeeze the last drop of b= andwidth out of the network you need a real Steiner tree for the set of nod= es being connected. Mind you if you are so concerned about bandwidth the 1= 0% or more overhead of the Bier header is likely also a concern. [Tony said] yes, but you don't provide any solution, just raise a problem. = Computing all the Steiners for all sets (which is combinatorial) is not fea= sible obviously. [PETER] Clarifying a problem actually. I raised it in the meeting but was t= old it was not a problem because MT solved it so I am simply clarifying tha= t it is still a problem and MT does not solve it. It may not be a critical = problem but it is none the less true that you could end up with some pretty= horrible trees interconnecting small numbers of nodes. The only solution that I know of along the lines of BIER would be to encode= the links to traverse as a bit set in a header as that would allow an arbi= trary tree per set of destinations. However for that you'd need to have a b= igger header. The alternatives to that are of course the hash based headers= that use Bloom filters but they produce false positives and unwanted traff= ic .. also possible loops. Unfortunately its trade-offs all the way down. I think BIER is actually a p= retty good set of trade-offs but let's be clear on what it can and cannot a= ccomplish. Peter --_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D5A2dfweml703chm_ Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Exactly, if you assign= a bit per adjacency and the header encodes the set of adjacencies to cross= then you get segment routing for multicast. I.e. source routed multicast a= nd can create any tree you want ingress. Pretty cool if you can afford the overhead. I suppose you could compress w= ith the header encoding logical adjacencies into some set of abstract topol= ogies .. probably give you similar behavior to the MT solution.<= /span>

 

Peter

 

From: Antoni P= rzygienda [mailto:antoni.przygienda@ericsson.com]
Sent: Tuesday, November 18, 2014 1:57 PM
To: AshwoodsmithPeter; bier@ietf.org
Subject: RE: Follow up on my comments from Honolulu concerning T.E/ = Steiner Tree optimization with Bier

 

Peter, yes, agreed, ve= ry fair summary.   I didn’t say in the meeting that MT solv= es it, must have been someone else ;-)  

I only clarified with = you that  I considered the ‘Steiner problem’ a single Stei= ner for _all_ I/E on the tree so we talked about different assumptio= ns.

 

I actually _can= _ imagine a variation of BIER encaps that would list the links but I think = that would be truly  ‘multicast segment routing’ rather th= an BIER

as was proposed in Haw= aii ;-)

 

tony

 

From: Ashwoods= mithPeter [mailto:Peter.As= hwoodSmith@huawei.com]
Sent: Tuesday, November 18, 2014 10:52 AM
To: Antoni Przygienda; bier@ietf.or= g
Subject: RE: Follow up on my comments from Honolulu concerning T.E/ = Steiner Tree optimization with Bier

 

Now does this matter? Well  if you are trying t= o squeeze the last drop of bandwidth out of the network you need a real Ste= iner tree for the set of nodes being connected.  Mind you if you are s= o concerned about bandwidth the 10% or more overhead of the Bier header is likely also a concern.

[Tony said] yes,= but you don’t provide any solution, just raise a problem. Computing = all the Steiners for all sets (which is combinatorial) is not feasible obvi= ously.

 

[PETER] Clarifying a p= roblem actually. I raised it in the meeting but was told it was not a probl= em because MT solved it so I am simply clarifying that it is still a proble= m and MT does not solve it. It may not be a critical problem but it is none the less true that you could end up w= ith some pretty horrible trees interconnecting small numbers of nodes.

 

The only solution that= I know of along the lines of BIER would be to encode the links to traverse= as a bit set in a header as that would allow an arbitrary tree per set of = destinations. However for that you’d need to have a bigger header. The alternatives to that are of course the h= ash based headers that use Bloom filters but they produce false positives a= nd unwanted traffic .. also possible loops.

 

Unfortunately its trad= e-offs all the way down. I think BIER is actually a pretty good set of trad= e-offs but let’s be clear on what it can and cannot accomplish.<= /o:p>

 

Peter

--_000_7AE6A4247B044C4ABE0A5B6BF427F8E20A43D5A2dfweml703chm_-- From nobody Tue Nov 18 11:54:42 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9A62A1A86F6 for ; Tue, 18 Nov 2014 11:54:40 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.795 X-Spam-Level: X-Spam-Status: No, score=-4.795 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.594, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id u1GhXhWPDf2t for ; Tue, 18 Nov 2014 11:54:39 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id CC1D81A8034 for ; Tue, 18 Nov 2014 11:54:38 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml404-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BOY18804; Tue, 18 Nov 2014 19:54:36 +0000 (GMT) Received: from DFWEML701-CHM.china.huawei.com (10.193.5.50) by lhreml404-hub.china.huawei.com (10.201.5.218) with Microsoft SMTP Server (TLS) id 14.3.158.1; Tue, 18 Nov 2014 19:54:34 +0000 Received: from DFWEML703-CHM.china.huawei.com ([10.193.5.130]) by dfweml701-chm ([10.193.5.50]) with mapi id 14.03.0158.001; Tue, 18 Nov 2014 11:54:27 -0800 From: AshwoodsmithPeter To: Antoni Przygienda , Eric C Rosen , "bier@ietf.org" Thread-Topic: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier Thread-Index: AdACs3lBaIcIqcEDSWSeZIwYsOpQZAA1gi8AAAmjE1D///J1AIAAgEFA Date: Tue, 18 Nov 2014 19:54:26 +0000 Message-ID: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D5B8@dfweml703-chm> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> <546B6938.9020709@juniper.net> <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D531@dfweml703-chm> <2E4BB27CAB87BF43B4207C0E55860F181E8BA5@eusaamb103.ericsson.se> In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F181E8BA5@eusaamb103.ericsson.se> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.193.60.150] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/ugnr9eIZQkvY836vUfEQFK8QzJc Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 19:54:40 -0000 Ok, but not one per sender as in 2201, that would be too many MT's, you'd l= ikely want one per 'corner' of the network so that if a cluster of members = are near that corner they use the tree rooted at that 'corner'. Cheers, Peter > -----Original Message----- > From: Antoni Przygienda [mailto:antoni.przygienda@ericsson.com] > Sent: Tuesday, November 18, 2014 2:31 PM > To: AshwoodsmithPeter; Eric C Rosen; bier@ietf.org > Subject: RE: [Bier] Follow up on my comments from Honolulu concerning > T.E/ Steiner Tree optimization with Bier >=20 > > > > Very likely what you want is to have a number of MT's where each MT > > has a different 'center of gravity' for want of a better word. When > > dealing with small sets of members that are localized, the centers of > > gravity of that set of members is used to pick an appropriate MT, > thus > > minimizing the horrific kinds of routing that would otherwise result. > > So a bit of offline analysis will be required to pick a good MT and > > MT's will need to be distributed properly to avoid bad degenerate > routing. > > >=20 > [Tony said] RFC2201 ;-) From nobody Tue Nov 18 12:02:40 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id ABD981A870A for ; Tue, 18 Nov 2014 12:02:38 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.201 X-Spam-Level: X-Spam-Status: No, score=-3.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, GB_SUMOF=1, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YGwzMF8wOdFI for ; Tue, 18 Nov 2014 12:02:36 -0800 (PST) Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 5AEC31A870D for ; Tue, 18 Nov 2014 12:02:33 -0800 (PST) X-AuditID: c618062d-f79206d0000014d2-f4-546b4cc797b0 Received: from EUSAAHC003.ericsson.se (Unknown_Domain [147.117.188.81]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id 3F.EA.05330.7CC4B645; Tue, 18 Nov 2014 14:42:32 +0100 (CET) Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC003.ericsson.se ([147.117.188.81]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 15:02:26 -0500 From: Albert Tian To: Antoni Przygienda , Ross Callon , "bier@ietf.org" Thread-Topic: BIER and deterministic ECMP Thread-Index: AQHQAruuIBpZwkXRcE2ekC470Sj+dpxmkeDwgAALgoA= Date: Tue, 18 Nov 2014 20:02:25 +0000 Message-ID: References: <2E4BB27CAB87BF43B4207C0E55860F181E8A95@eusaamb103.ericsson.se> In-Reply-To: <2E4BB27CAB87BF43B4207C0E55860F181E8A95@eusaamb103.ericsson.se> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.4.140807 x-originating-ip: [147.117.188.10] Content-Type: multipart/mixed; boundary="_002_D090DCA031617AAlbertTianericssoncom_" MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFjrNIsWRmVeSWpSXmKPExsUyuXRPoO4Jn+wQg+Z7/BZLZ+xhsli34QOz xas5d5gt/q64wmKxb9E8RgdWjym/N7J6LFnyk8njetNV9gDmKC6blNSczLLUIn27BK6MuVPf sxYsOM5ccX7bZtYGxnvbmLsYOTkkBEwknp09xAhhi0lcuLeeDcQWEjjCKLF1r3oXIxeQvZxR 4ub0LWBFbAI6Eq9mPwCzRQRqJKYfWcICUsQssJZR4uPmU2BThQU0JPbf3sAMUaQp0fR8GpDN AWRbSayalQkSZhFQlXjffwtsDq+AmcT6xgYmiGWzGCWalx5kBUlwCvhIzL5+lQXEZgS67vup NUwgNrOAuMStJ/OZIK4WkXh48TQbhC0q8fLxP7BeUQE9iWcbNrNDxJUkJi09xwrRGyox59oN ZojFghInZz5hmcAoNgvJ2FlIymYhKYOI60ncmDqFDcK2lti7/iEzhK0tsWzhayhbX2L+1COM s4BeZhaQk5i2xX0WOIiA4Xj3+Q9WmN4nq7uh5ihKTOl+yL6AkXsVI0dpcWpZbrqRwSZGYOwf k2DT3cG456XlIUYBDkYlHt4NrlkhQqyJZcWVuYcYpTlYlMR5Z9XOCxYSSE8sSc1OTS1ILYov Ks1JLT7EyMTBKdXAKD7FX8mnjD9gTvS8voq/pyXS+5TNluzJ8XnTtWv6i9ySAJ1mVjlL571O y/+yLwj++mmN+7/Nc9gnJ0T4REtPCt9mHiw4wzT84t2VE6JitDZnlJ85zOQxSUL98c9Pwdvu 9j7N7C+q6j6xQtPi5/9/hwq+BovpTqs/6b5KhffsGbcJzyfMqxZoU2Ipzkg01GIuKk4EALQd r8neAgAA Archived-At: http://mailarchive.ietf.org/arch/msg/bier/_mtaAyOM0KEnfmbf9sGLnRnWANk Cc: "Greg Shepherd \(shep\) \(shep@cisco.com\)" , "IJsbrand Wijnands \(ice@cisco.com\)" , "Eric Rosen \(erosen@juniper.net\)" Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 20:02:38 -0000 --_002_D090DCA031617AAlbertTianericssoncom_ Content-Type: text/plain; charset="iso-8859-1" Content-ID: <0F803D72ED0E424CADDE57960B76F336@ericsson.com> Content-Transfer-Encoding: quoted-printable Nice picture Tony :-) Given the limitation of BIER, some heuristics such as defer the split (as you alluded to) can reduce the total cost of the tree. For example, any BFR can try to locally minimize the number of copies (or the sum of the oif costs to be precise). However in some examples, like the one in the attached topology, the global tree optimization is hard to achieve in the current BIER. In this example, A is trying to send to X Y Z. The best tree is to group Y with X (compared to grouping Y with Z). A needs to make that decision, however from its BIFT, A would only know Y has ECMP next hops B and C, and they are equally good from A=B9s point of view. On the topic of what behavior do we want: please note that receiver may see out of order packets due to path shifts, which are undesirable. Trouble shooting non-deterministic paths could also be a headache. Thanks, Albert On 11/18/14, 9:41 AM, "Antoni Przygienda" wrote: >> Reading the BIER architecture (draft-wijnands-bier-architecture-01) I >>noticed >> the last sentence of section 6.7: >>=20 >> Note however that by the rules of Section 6.5, any packet destined >> for both BFER-D and BFER-F will be sent via BFR-C. >>=20 >> Thinking about this more: It is clear that where ECMP paths exist, >>which path >> is taken to reach any particular destination will depend upon which >>other >> destinations the same packet goes to. In particular, which lower >>numbered >> bits are set in the BitString can change the choice among ECMP paths. >>This >> could have the effect that as a lower numbered destination >>intermittently >> joins and leaves a multicast group, the path to a higher numbered >> destination could switch back and forth between different ECMP paths. > > [Tony said] Ross, I thought more about it & it's not a simple ECMP >problem in my eyes. What you suggest [entry per destination] can lead to >quite detrimental effects under normal operation since we're operating >here in the domain of fan-out counts as well (I'll explain) [unless I >misunderstood your email]. > >Observe the attached figure (I hope it survives the mailing list send >for others). When you're sending to a XYZ you clearly prefer the >upper-most path due 1 replication count (latest possible fan-out). >Tables split per destination can lead to twice the bandwidth network >consumption on the network if A ends up replicating along ACX, ADY, AEZ >(immediate fan-out). When X falls off the receiver list, AD is the >prefered 'axis' and if only Z is receiver, you have 3 ECMP paths all >consuming same amount of bandwidth. > >My point: what is the property we try to preserve ? Given the example, >I would argue first that changing receiver list you may want as >preference to change the paths through the network since the optimal >replication points may shift to save bandwidth. Using ECMP too early can >lead to unnecessary duplications. As long the receiver list is the same >however, something like entropy label should control the ECMP. > >I am not saying the problem does not exist, I am just introducing another >important axis that can deteriorate when looking @ optimizations in >another axis.=20 > >--- tony=20 > > --_002_D090DCA031617AAlbertTianericssoncom_ Content-Type: application/pdf; name="topo1.pdf" Content-Description: topo1.pdf Content-Disposition: attachment; filename="topo1.pdf"; size=39682; creation-date="Tue, 18 Nov 2014 20:02:25 GMT"; modification-date="Tue, 18 Nov 2014 20:02:25 GMT" Content-ID: Content-Transfer-Encoding: base64 JVBERi0xLjMKJcTl8uXrp/Og0MTGCjQgMCBvYmoKPDwgL0xlbmd0aCA1IDAgUiAvRmlsdGVyIC9G bGF0ZURlY29kZSA+PgpzdHJlYW0KeAGtl8luG0cQhu98ioqzeAHc6qrer1FyyC22COTii0HQsAwp ipbEr5+/ZqY5PcMhKQkiL2SBU91fLX8Vb+kD3ZKLxiZJ0n2w1golLqbgRcFbutvSX/Q3nZ3fM23u ibv3/WZ8LKTQ/Xp1TSl0viJVG12NNj0HLwfb/Ei1faUv76gECfR9xfStP/D8gmz3vjjHTW3nIOXk LfvO/H5qYvGGUwCKyyYHiavNNelvIhd2keFBrA/MkXBZb3PxpjRG3LYaXSzOifG4WeNg1Zi/0gXi h7iIxqW/JuJyiwDpFwTKsXH9bYIR3IZwm1/XnT9r/Wq9Ic79bzNxNCWVkug9bOtrOluvBT7WX+jN D6/e0vob/b7u0jUl1pNmQXA5GcSfxHuDULHGAA/hq8seYTO2SGIvMsYAtxituyBk5KkLAiMIo4ca A7X2MRiRHQKaurPZCM4eiPGwtYGmwNnYIOIUeDUF/vFpwII8IZNxQoygcIy5iOb3WcSjh2PE4tV9 d/g+ssyQk3E5x7iA/NPTkGuOkVKDbpWlHLPlhRz31v0cy2KO1Xoox94Fwzi7zfEcOJoYrMsLwD8/ DVicNYGdpxmxQzjZhz7HE2JcrrUOxGodqnogHjzUHC8Ri0uIcnf4BLlvPDTppJGTkcJlifmXpzHX JLdiNrYh4Pa0TPu4GvdTrFo2Pl+B1XooxQ4pboQLD6t+z3ETmiwspPh1g3v25/Zus/3n4d/PV3R3 CZEMxYiLLvb6Z8WwLZDMlPsPAiXPLnot7LM/rpl+u4H6CQaSpe909nF79fnh8r/t+c3Vzd3l9fbh 7nKjfq1xtmTrtOmDDTmLCh6EiFMMq0cOkSx9qTncwXFX35rpEGzOYSi1XGLImHjRlQyZQx+4IAmh 7GbCHJYtZFGS28F6oJFAqG2xhRQ2AH+AlR72IOTtXPGPDIGBQT0rw2QIdAwYApAJF6BgJyEwwPXe elyBunbpk4yWUDTPyWCwuAHCnYToqskemeaSdYhjVvturOegrvGUDz7HwAqEEVs8JotCBAdfxLij TakcSgU2lAIHSCPebDF4a8k5hpYCagDwzwCYIvGskGoSaiFpQorULERbQsHeBZKYnY2HANg64+F6 ngZIlEd1ai1xxkjtGyecpNBb4HUkD1WHVIMdJKFyNMXkEholpLaY4DUKqqpviX6x6+5s6XF9GCA+ PqM7mj0Gl5UQkKTJxjLZ5XazfieApSq+biyjhyqAS3tM8GKwPODs/aG+L4AF9Et7zKc3jQI+Yn2t xK3gj/dF6VdtnwBX4z7vsLwOEau8S4JfeR8n+Nly8guC/+nt83jbmT7hRS4nM73f1nfWfWKd3qOH Srw00yvxbItpFnJs0Yy/DP0+vmr28XcN5d5cKxFLd5X6nUp61JNo085UMvbt+RJzDdgn+riWV5VT XdzGfsLzg5pirDG7FFHTp9T0BK1Kqs71QYzSC4hRhWgG28gwEdSRQRsAY+TAcD7BMBPU/IIMraBO IBo5HRPRyul+2SHQUP1+KmCsGdfNZMiXqiVpInTnGBJRngMxra+jedBaHCfbLhGTwfbhfxvIk9YK ZW5kc3RyZWFtCmVuZG9iago1IDAgb2JqCjExNzUKZW5kb2JqCjIgMCBvYmoKPDwgL1R5cGUgL1Bh Z2UgL1BhcmVudCAzIDAgUiAvUmVzb3VyY2VzIDYgMCBSIC9Db250ZW50cyA0IDAgUiAvTWVkaWFC b3ggWzAgMCA3OTIgNjEyXQo+PgplbmRvYmoKNiAwIG9iago8PCAvUHJvY1NldCBbIC9QREYgL1Rl eHQgL0ltYWdlQiAvSW1hZ2VDIC9JbWFnZUkgXSAvQ29sb3JTcGFjZSA8PCAvQ3MyIDggMCBSCi9D czEgNyAwIFIgPj4gL0ZvbnQgPDwgL1RUMiAxMCAwIFIgPj4gL1hPYmplY3QgPDwgL0ltNiAyMSAw IFIgL0ltMyAxNSAwIFIKL0ltNCAxNyAwIFIgL0ltMiAxMyAwIFIgL0ltNSAxOSAwIFIgL0ltOSAy NyAwIFIgL0ltMSAxMSAwIFIgL0ltNyAyMyAwIFIgL0ltOAoyNSAwIFIgPj4gPj4KZW5kb2JqCjIx IDAgb2JqCjw8IC9MZW5ndGggMjIgMCBSIC9UeXBlIC9YT2JqZWN0IC9TdWJ0eXBlIC9JbWFnZSAv V2lkdGggMjk2IC9IZWlnaHQgMjggL0ludGVycG9sYXRlCnRydWUgL0NvbG9yU3BhY2UgNyAwIFIg L1NNYXNrIDI5IDAgUiAvQml0c1BlckNvbXBvbmVudCA4IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlCj4+ CnN0cmVhbQp4Ae3QMQEAAADCoPVPbQo/iEBhwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGPgODGEgAAEKZW5kc3RyZWFtCmVuZG9iagoyMiAwIG9iagoxMzIK ZW5kb2JqCjE1IDAgb2JqCjw8IC9MZW5ndGggMTYgMCBSIC9UeXBlIC9YT2JqZWN0IC9TdWJ0eXBl IC9JbWFnZSAvV2lkdGggMzI0IC9IZWlnaHQgMjggL0ludGVycG9sYXRlCnRydWUgL0NvbG9yU3Bh Y2UgNyAwIFIgL1NNYXNrIDMxIDAgUiAvQml0c1BlckNvbXBvbmVudCA4IC9GaWx0ZXIgL0ZsYXRl RGVjb2RlCj4+CnN0cmVhbQp4Ae3QMQEAAADCoPVPbQwfiEBhwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDDwHBhqUAABCmVuZHN0cmVhbQpl bmRvYmoKMTYgMCBvYmoKMTQyCmVuZG9iagoxNyAwIG9iago8PCAvTGVuZ3RoIDE4IDAgUiAvVHlw ZSAvWE9iamVjdCAvU3VidHlwZSAvSW1hZ2UgL1dpZHRoIDE3NCAvSGVpZ2h0IDMzMCAvSW50ZXJw b2xhdGUKdHJ1ZSAvQ29sb3JTcGFjZSA3IDAgUiAvU01hc2sgMzMgMCBSIC9CaXRzUGVyQ29tcG9u ZW50IDggL0ZpbHRlciAvRmxhdGVEZWNvZGUKPj4Kc3RyZWFtCngB7dABDQAAAMKg909tDjeIQGHA gAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAED BgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwY MGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDA gAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAED BgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwY MGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDA gAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAED BgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwY MGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDA gAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAED BgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwY MGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDA gAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAEDBgwYMGDAgAED BgwYMGDAwNPAAKECAAEKZW5kc3RyZWFtCmVuZG9iagoxOCAwIG9iago3NzUKZW5kb2JqCjEzIDAg b2JqCjw8IC9MZW5ndGggMTQgMCBSIC9UeXBlIC9YT2JqZWN0IC9TdWJ0eXBlIC9JbWFnZSAvV2lk dGggMzIwIC9IZWlnaHQgMzEzIC9JbnRlcnBvbGF0ZQp0cnVlIC9Db2xvclNwYWNlIDcgMCBSIC9T TWFzayAzNSAwIFIgL0JpdHNQZXJDb21wb25lbnQgOCAvRmlsdGVyIC9GbGF0ZURlY29kZQo+Pgpz dHJlYW0KeAHt0AENAAAAwqD3T20ON4hAYcCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAgY+BAZX8AAEKZW5kc3RyZWFtCmVuZG9iagoxNCAwIG9iagox MzM0CmVuZG9iagoxOSAwIG9iago8PCAvTGVuZ3RoIDIwIDAgUiAvVHlwZSAvWE9iamVjdCAvU3Vi dHlwZSAvSW1hZ2UgL1dpZHRoIDMxNiAvSGVpZ2h0IDI4IC9JbnRlcnBvbGF0ZQp0cnVlIC9Db2xv clNwYWNlIDcgMCBSIC9TTWFzayAzNyAwIFIgL0JpdHNQZXJDb21wb25lbnQgOCAvRmlsdGVyIC9G bGF0ZURlY29kZQo+PgpzdHJlYW0KeAHt0AENAAAAwqD3T20PBxEoDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMDA/8AAZ7AAAQplbmRzdHJlYW0K ZW5kb2JqCjIwIDAgb2JqCjEzOQplbmRvYmoKMjcgMCBvYmoKPDwgL0xlbmd0aCAyOCAwIFIgL1R5 cGUgL1hPYmplY3QgL1N1YnR5cGUgL0ltYWdlIC9XaWR0aCAzMDAgL0hlaWdodCAzMzEgL0ludGVy cG9sYXRlCnRydWUgL0NvbG9yU3BhY2UgNyAwIFIgL1NNYXNrIDM5IDAgUiAvQml0c1BlckNvbXBv bmVudCA4IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlCj4+CnN0cmVhbQp4Ae3QAQ0AAADCoPdPbQ43iEBh wIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBg wIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBg wIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBg wIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBg wIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBg wIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBg wIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYM GDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBg wIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIAB AwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBgwIABAwYMGDBg4GNgi+gAAQplbmRzdHJl YW0KZW5kb2JqCjI4IDAgb2JqCjEzMjIKZW5kb2JqCjExIDAgb2JqCjw8IC9MZW5ndGggMTIgMCBS IC9UeXBlIC9YT2JqZWN0IC9TdWJ0eXBlIC9JbWFnZSAvV2lkdGggMTgxIC9IZWlnaHQgMzEyIC9J bnRlcnBvbGF0ZQp0cnVlIC9Db2xvclNwYWNlIDcgMCBSIC9TTWFzayA0MSAwIFIgL0JpdHNQZXJD b21wb25lbnQgOCAvRmlsdGVyIC9GbGF0ZURlY29kZQo+PgpzdHJlYW0KeAHt0AENAAAAwqD3T20O N4hAYcCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCA AQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBj4 GBiV5gABCmVuZHN0cmVhbQplbmRvYmoKMTIgMCBvYmoKNzYyCmVuZG9iagoyMyAwIG9iago8PCAv TGVuZ3RoIDI0IDAgUiAvVHlwZSAvWE9iamVjdCAvU3VidHlwZSAvSW1hZ2UgL1dpZHRoIDI5NiAv SGVpZ2h0IDI4IC9JbnRlcnBvbGF0ZQp0cnVlIC9Db2xvclNwYWNlIDcgMCBSIC9TTWFzayA0MyAw IFIgL0JpdHNQZXJDb21wb25lbnQgOCAvRmlsdGVyIC9GbGF0ZURlY29kZQo+PgpzdHJlYW0KeAHt 0DEBAAAAwqD1T20KP4hAYcCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBj4DgxhIAABCmVuZHN0cmVhbQplbmRvYmoKMjQgMCBvYmoKMTMyCmVuZG9iagoyNSAw IG9iago8PCAvTGVuZ3RoIDI2IDAgUiAvVHlwZSAvWE9iamVjdCAvU3VidHlwZSAvSW1hZ2UgL1dp ZHRoIDI5NiAvSGVpZ2h0IDI4IC9JbnRlcnBvbGF0ZQp0cnVlIC9Db2xvclNwYWNlIDcgMCBSIC9T TWFzayA0NSAwIFIgL0JpdHNQZXJDb21wb25lbnQgOCAvRmlsdGVyIC9GbGF0ZURlY29kZQo+Pgpz dHJlYW0KeAHt0DEBAAAAwqD1T20KP4hAYcCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMG DBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgwYMCAAQMGDBgw YMCAAQMGDBgwYMCAAQMGDBj4DgxhIAABCmVuZHN0cmVhbQplbmRvYmoKMjYgMCBvYmoKMTMyCmVu ZG9iagozOSAwIG9iago8PCAvTGVuZ3RoIDQwIDAgUiAvVHlwZSAvWE9iamVjdCAvU3VidHlwZSAv SW1hZ2UgL1dpZHRoIDMwMCAvSGVpZ2h0IDMzMSAvQ29sb3JTcGFjZQovRGV2aWNlR3JheSAvSW50 ZXJwb2xhdGUgdHJ1ZSAvQml0c1BlckNvbXBvbmVudCA4IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+ CnN0cmVhbQp4Ae2d13IjxxJERQNHB3oStAuaXXrv/f9/lcp097SpBvfpRugy+YLpGihCOgK5ydqc zH/+wRcIgAAI/AcJjPHXf/Df+3/+ryygQOtvuI+NjY9P8Nf4OD5bo4ERqonJVqvdak1OANZIVkKq 0+31et1Oa5I+WiPf/aNv0qdqst2dnp3rz81Od9v80frRPEb9xzOq3sz80srqytL87FRncgI/syq8 xsYmWt2Z+ZXB1vb2xtrSHGBVQNGYflq1p/rLm78Ofh/u7w6W+wQLP+BtXvQt2JlZWN89PD0/P/0z 3CBYbcCqsurOLm3snVzd3t1cHBGsOcCyUdH34CSx2tw/vXl8frq/PB5uLM318Mkyacn34OLG3unt 09vb68PV0XCwNAtYFVb0s31+7dfR1ePbx8frw+XRr8HibLcFmVXiYs3Qm1vePDi7ffn4/Hi9v/yz u744A1glKhYNk53p+bWd3xf3r59fBOvi9+7aAmCZrESMLq7LdyHBerk7P9xZnZ+mTxYEfA6M1WiP ZMPw+PrxTWCdHWwTLPy2k5Oib0L6Lmzzj6y9k5un96+vj+fb04OtFQj4EpWDxb/nkMh6Jljvzzen +5sECzKrxMWfrM5Uf2WL/jB8/vj6fH+6PtnbXIYmLVHxJ2uC/zBc3T48u3shWG+P1yzgoUkrsFoM a+fw3MFiAb8IWCascV5jLaztqsz6FAG/DgFvseI9FsFaXN/9c/lAmpRgXZCAhyatwWp3Z1mTXj6w zGIBvwMBb7LiX3ZYkw6Gx/RrtBPwokkh4AtgTpOSgD+5fnonWM+3LOD7JOCxVM5pCawpEvD7KuDf ScDvQ8DnmPSssEiTegH/dHOyT5oUAt7gJQJ+ur+aCHjawGNPasLyAt5rUgh4A5P/NlRNyqs//W3n SpbKvTaWygU01aQk4P/InvTz7QFL5QKSH3gB/8sLeCyVPZryVfeki4OwVL6npfIalsolKb8nTZfK h1gqW6g8LFkqk4CXpfIZlso2K93Ai4A/lQ08lsoVUDwWTcpL5f1TXip/vZOA56UyBLwBTQU8LZXl r6NlA3+8B1eIQUo+WbmAvzqGK8Rm5fakC/p39/TXFcEVAgFfEnMCnpfKbHRgWHCFlJh0IrCypTJc IRVaulReHAzZm+WWynCF2LBk9QdXiA0nnyqsRMDDFZJD8meBBVeIxzH6VTRp4wohAQ9XSJWYwCpd IdjAW8TYe1u6QmChsVgFAS+uEPYbsa0brhCTFVwhNhZ7yktluEJsNsU0uEIiAQ9XSIFJBk6Tiq1b XCEvd97WDVdIjkw1qQj4YOuGKySn5M4KC66QCp50rJo02LpVwMMVkkLyJ9i6PYm/eGUB352mpfLh OVwh3/FyS+Vg64YrZAQxhRXZuuEKqdNiAd+bDa6QzxdxhSzgUUMDmWpSsXWHRw3hCjFA8UhhzXGc Q+IKga3bAKaalAR8sHXjUUMDk45Ek8IVUuWT3BBYvFSGKyThYh4aAe9t3XCFmKB46DWpJDr4pTJn hcAVUjJzsMgVorZuuEJKRmEisNgVQrZu/6ghXCEBT3rhXSHB1o2skBRQdFJNClt3hKR+qbASVwhs 3TVcAsu5QtgDD1t3jRTNRZOKK8TbuuEKqeISWCrg46wQuEIsYnCFWFQqM1PAkyukh7C/kpjCipbK YutGVkhJiiYEC64Qk4wxrLlCEPZXwnKa1HKFIBkxx6WaNHWF4LnMnJI7K6z+Sgj7e8IGvoLKCfjp zBWCsD+bV7NURtifTSiaRq6QEPbHAdSSFYKij4gUXXoB78P+nCsEad0pJj2VAh4B1BYnmbGAj1wh FPZ37sP+EECdUVNNmrhCztUVgrC/DJUqBwqgdq4QF/a3hbC/AhQPgoCPXSEc9ocA6pIXw5IAamSF lHDyicCisD92hbi0bsoKQdhfzknOjYCHK8QEFA+9Jm1cIQj7i/kk1w4WXCEJlcpBYLmsEDTIVCD5 sXeFJFkhCPvzeJJXL+DjBhnYuhNEzUFhwRXSEBlxJbDgChlBKLolmjRxhSDsL8KTXgqs2BXyJAHU cIWkmPQEV4hFpTIrBbxWQMIVYgBTWOIKQdifwScZWQIeDTIJouZguUIQ9tfwia+cJkUFZAyldq2a FK6QGp9krrBQAZlAqR1EkyLsr4YnnTdLZbhCUjLGyRLw6gpBRViBywt47wrhRw0R9ldg0kEk4FEB WWEUxgSLbd3UIHMpAdSogAxoigvVpKiALMBYA4WVuUJQAWmhslwhqIC0SdHUCXhUQFYJRTdEwDcV kAj7i9gUl42A964QVEAWkPyABTzXku/8RtifZ1J9DQIeWSFVRuGGwHJZIVK3g7C/wKa48K6QkBWC sL+CkR94AQ9XiCcy4tUJ+OVNhP2NoORuqSalDTzVkiMr5BteoknhCvmGUvTJ4lKUOOxviApIG14p 4BH2Z5Oiqdekja3buUIQ9lcyc7BQAVmiKSdBwB81YX9whZScZOIFfGLrXkWHu4XLEvAH6HC3ULk9 aS+xdaMC0kYVlsoi4J857A9ZITVUCkuey9RHDVEBWUfFsHwteeIKga3bgsZLZVRAWmSMmRfwqIA0 4OQjhRWH/V1QhzvC/nJOciZYCPszyRhDy9aNCkgDFI3cUtkK+0MFZI4sLJX3TlABmcPJzwoLFZA5 F/PMsDp52B+yQkxWNQFP+XVwKhfEgiuEKyA/qZb8EVkhBSQ/SAQ8wUKDjCdjvAoscoXsUoMMXCEG oHjEAj4K+0MFZAwnu1ZNmoT9nSErJIPkjwrL27q/vj6eb6VBBhWQnlD0qpoUrpAISf1SNCm7QhD2 V4fk7wisKCvkHVkhHk352mzgva0brpCSkpt4TQpXSBVRc8PBQthfg6R+5QU8PWqICsg6Jr0DV8h3 hJr7qklRAdkQGXGlsBJXCCoga7wEFsL+anjSuWjSxNaNCsiUUHQKmvQArpAIi30JV4jNxZyaAp4q IKl7bhxtOxkyhRUtlVEBmRGKjgQLrpCIx8hLuEJG4kluOk2qrpA3apB5oaUybN0Jo3BQTSoCHq6Q QKVy4QR8XAF5sk+thqjbMYCJgIcrxCBjjIKAP0wEPFwhBiu20LS605QVchg28EehAtL6B37yzAv4 OOxvFwLe/kgUAh5ZITYonrKAz1whqICs4FJNmrhCXAXkJGrJc2YKK3eFoAIy5yRn1aTsClFb9/sz 1ZKjAtJkJQ8MhApIfS7zZA8CfgSsogJyCY8aWrgaAR806fFwsAQBb8Iab8L+Xj/I1v0AW7fFSWZO wMMVUiUU3RBYTQXkJwdQ767hucwIUXNpukJQAdkAiq68gEfYXwSldukEfBH2N9XBM04FMxXwaJAp wFgDhqVZIaeulhwVkBYnmQVNCldIlVG4wUtlZIUEHKMvnCaltO6L+xcS8PSooWzg4QoxuCksVEAa aMpRJuA/0CBTMgoTyxXCFZCdFpbKAZK7cJoUFZA5GOusmhSuEItNMbMF/HIfrpACFcL+DCT1URDw cIXUIfk7loAPFZD+TXhVAl7Ax64QVEBWPh2RgPdZIQj7q7ASV4hWQF6hArIGyc1Vk1oVkHCFFOiC gJcGmWDr7iMrpEDlkhGnGlcIBauoKwQb+JIWf7KcK+TGpXXD1l1i0onAaiogPykr5GQPYX82rkbA e1cIKiBtUjT1mpQrIMUVgrC/KqsAaxcVkHVI/k5YKiPszyOpv3pXSFoBiQYZi5gX8HCFWHSymRPw cIVkXMyjwHJZIaJJ2da9uYKlskVLNGmTFUKaFK4Qi5PMBFYU9veGsL8qK3nUUPoym0cNYeuu4WoE fOYKQQVkiczBQgVkiaacBAEfhf2hArLkJBO4QipgjLHTpHCFGGyKkWpScYVccy05L5UP+FFDLJUL VvKoYTsW8FIBCVdISYomjYA/Y1s3KiBNSm7YLJWRFTKKk9yzXCHDwaI+l/ntP/3D3uAFPCog/+J/ vMKKwv7uUQFZxUawOOxv4NO6X+i5TAh4G5dqUrhCbDrZNBbwTxL2d+fD/lABmbFymtTbuslCA1dI jqg56287qIBsiIy4EgGPsL8RhKJbEPARjO8uWcB3ZyTsDxWQ38JywSqxrRthfxVqfqmMCsgKoHjM Aj4K+/t4uT8/JAE/TU/8jiEHPiblzLc9NMikVConJ+CXN8TWjQrICiYdqyZtbN0S9gdXiM1MNCkq IG04+VQF/Pzq9sHZLSc6wBWSE4rOjYD3tm5UQEZ40kvdk5KAp6wQDftzWSFwhaSg+ORgIeyvRFNO vID3S2WE/ZWMwsTbuodH+lzmyx0JeGSFBD7xhWpSCPiYSfVaYSWuEFRA1mgJrKICEq4Qk5do0sbW DVeISckNgyZNw/4QQG1B46VyGfYHV4jFymvSVMCjQcZk5WDFrpBLZIXYqOS3HbhCanCyudekg+Fx I+DRaphRckenSVEBaeNJp6pJ4QpJqVROCquPCsgKn2TMsDpwhSRMqocg4BH2V2UUbrCARwVkwDH6 wm/g4QoZzUnuKqxYwKNBpoqNYMWukE9yhXhbN1whGTXVpKiAzLDYR4WFCkibTjZVTZq5QlABmVFy R9Gk3hWCCkgbkp8KrKgCkl0hexuogPR8ktdGwMMVkoCxDl6TNq4QVEBanGTmYMEVUiUU3RBYqICM iIy4NF0hVAGJBpkSmmpSuEJKMsZEYcEVYqApRwIrzgqRsD+4QkpSNBFN2rhCEPZnUnLDoElTVwg6 3C1ovFQuXSHc4T6BZ8lzYLGADxWQcIXkmPSssKIKyCYrBHvSHJkT8JIV8so9H/So4faqJt/m7/3p Z69JB2rr/qREh7P9zeW5Hh7LLD8aTpP6sL/Pt8fro1/rC1IgU777h09Uk6qAf3z7eH+hXJXNpdlu axyP+xYfDYUlHe7Xjy+vz5RutL0i34TFWzFgWFq3c3J9//h4e06s+vwDC2hKAl7Ab+0fX1xfX57s b+kP9/KdmNCvhirgt4a/j0+ODyn5YgY/r2qfC5FZ0/MrGzvDveHOBm0bOMa19u4fPldY/aXVwcZg bVnKUMCq9plgWJ2p2f7C4uLC3HS3jY9VjZRssyZa7e7UzMzMdK/TIlT4FqzSIuUwMdlqd+ir3cJO pspJbowxrYlJ+poAqtGo6C7Tki+6+PbNP/4NBEm+fjwIAAABEACB/28C/wKhi7E1CmVuZHN0cmVh bQplbmRvYmoKNDAgMCBvYmoKMzg1OQplbmRvYmoKMzEgMCBvYmoKPDwgL0xlbmd0aCAzMiAwIFIg L1R5cGUgL1hPYmplY3QgL1N1YnR5cGUgL0ltYWdlIC9XaWR0aCAzMjQgL0hlaWdodCAyOCAvQ29s b3JTcGFjZQovRGV2aWNlR3JheSAvSW50ZXJwb2xhdGUgdHJ1ZSAvQml0c1BlckNvbXBvbmVudCA4 IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4Ae3YWXeaQBQHcJCdGVmVJRAFFaNQl8QN 45J8/0/VGWgtJO2YPrbn/h84Rw++/LyXO1yOg4AACPxnAjzkbwR+8++Tn3doBMg9AcpEtT8oUkFB EEUJ8hUBURQoYwuRCoqSrKga5AsCqiJLRLGNSAllVcddwzQtCFPANE0DI02R2og8L4iKhk2n5/l+ AGEK+L7Xdy0DqW1EviPKWtf2wniQpOkIwhBI02T4GPmuiUglNrqZFyQV236cZrO8+AZhChT5fDoe hD1Dl8XObayQp6Gsm/14NFusN9vdHsIQ2G03L8s8GwQ2VojhD0QylEUF2cHwabHZH0/nC4QhcD69 ltvVfBS5XVVqG2InTOer3fF8fXuHMATerpfT4aWYxD1D+/VArOoQ20EyW27L0+UKYQpczq/752JM DMlk/tnLHB3LyPIeJ/l6sz+URwhDoCwPu+fFUxI6uNHLHDkeSprhhsPJrFiu1hCmwGq5yKdp1Lf0 xkzhaCEqyOwFcTKaZFMIUyDLxungwbNJGZI3ldsrMzncSCoynL4fRlEMYQpE0UPguRbW6PGwYUgK UVJ0bFi247gQtoDjWGYXqTLdOtzKkDQzrURZ1XSEEIawBRDSNVWpFjcNwhqRbr9kWYHcESBIUr36 apYh4azW2PUCV4QwBGqjapPdqsLqA91uk56G3BeoqT4TNhzrW+DKEPiDH3wNAiDwLwt8B3RcuqwK ZW5kc3RyZWFtCmVuZG9iagozMiAwIG9iago1MzcKZW5kb2JqCjI5IDAgb2JqCjw8IC9MZW5ndGgg MzAgMCBSIC9UeXBlIC9YT2JqZWN0IC9TdWJ0eXBlIC9JbWFnZSAvV2lkdGggMjk2IC9IZWlnaHQg MjggL0NvbG9yU3BhY2UKL0RldmljZUdyYXkgL0ludGVycG9sYXRlIHRydWUgL0JpdHNQZXJDb21w b25lbnQgOCAvRmlsdGVyIC9GbGF0ZURlY29kZSA+PgpzdHJlYW0KeAHtmNl6qjAURrWQMITJoAzi AEWsonWoOA/t+z/VSUA46tf7XrDXDfDB1eLf2cluNAAwAAb+0EATeDHwy8/gX7xxBIAb4Cq4kxdV 3JIgiCICSgOiyGS9iOKWRIQlWQHuBmQJI2bqWRTThLCsarphmhZgmaZhaESR0JMoniYkqbpF247r eoDnuk7HtnRVfhLFPImSalDH7w/DKHqvPVEUDvuBa5tEEoX/lddsCkjRqduPkkk6mwPzWToZxwPf NlQsvlU9j8UJq6bTiyef60223dWebbZZL9LRwG1pEvN0F5WXndbywo/FZn88X66153I+HbLl9L1r 6zJ69IQkjXaj6Wp3uty+f2rP9+162n/NRr22oaBqgSryRP1outwez9cbcLtejrt1Ggdtg3W8su4a vN0Ryx0ks1W22x+Otedw2Gfrz4/Qp9pD3TFPAlYM2x+OpvPFcgWslot5Oo6CjpX3u7LhsX0BC5TZ 9vthPErGwDhJ4mjQ7bRYnNiOvDwL5/txmZi04/lB0AN6QdD1HNvSFL59qjw1+EqO2PHOsFqU2oBt 05Zl6kTG/CRcxold80RhWVEJIRqgaYSoiizlA4MHTYUoNn3CGEsAM8BEoGKs8hinPFFsnFmMMsXa U3jIJ5pPacof+JST1R+QGyhsvISpknZ/CxduoLICN2AADPyNgX8JiHn3CmVuZHN0cmVhbQplbmRv YmoKMzAgMCBvYmoKNTE2CmVuZG9iagozNyAwIG9iago8PCAvTGVuZ3RoIDM4IDAgUiAvVHlwZSAv WE9iamVjdCAvU3VidHlwZSAvSW1hZ2UgL1dpZHRoIDMxNiAvSGVpZ2h0IDI4IC9Db2xvclNwYWNl Ci9EZXZpY2VHcmF5IC9JbnRlcnBvbGF0ZSB0cnVlIC9CaXRzUGVyQ29tcG9uZW50IDggL0ZpbHRl ciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngB7ZjZcoJAEEURGAZZRFliXMCoRImJiYJVKiYm//9V mRmFIA7ybKXvE9WUL8du+vYVBBAQAAL3SqABqiXA/W/pr0RQDQFKqYyPgZMkGXSTgCSJYpkeaTlJ RoqCsQqqIoAxVpDM6BVaj6JDiqrphgmqJmAYWlNVCL3C3DJ0qmZaHcd1PVAFAde12y2jiRGBlzce 7TpVt+yH3sAPgmAE4hAIAn/Yf/Q6poblQuM1RBlrltvzx+Esil5AXALRfBY+Dbu22Sw0Hms70+6N wsVytY4TEJdAvP54i6Z+t60XGo+y0yzPny5Wm+0+TQ8gDoE03W+T92jcd0wV5UNL2ClauzuaLTe7 z+PxG8QlcPw6bOPXcOi2VCRlm7bM7gfEIfD9x4588C7YnWd2t4eR5QwsLZ1mdnI1s0g1ndOuiGFX cDdFksTr1fWuEHKPMnmeg0fhGhRSjKLMoygFc8xMim45J2/M8YVQIgSIOR4Qb9wqeWPSeOh8k1Xc I1D2PHKTdazyTSawgxZnWUALxCFAIwJDv8oCBAYPMqiq8CmrswzqMkahmQDpPFGC7PNm8Ele8rJP Co9FxzWR879/TSnl+VPhgdZBNQQKvOARCACBOyPwC09gpooKZW5kc3RyZWFtCmVuZG9iagozOCAw IG9iago0NjYKZW5kb2JqCjQxIDAgb2JqCjw8IC9MZW5ndGggNDIgMCBSIC9UeXBlIC9YT2JqZWN0 IC9TdWJ0eXBlIC9JbWFnZSAvV2lkdGggMTgxIC9IZWlnaHQgMzEyIC9Db2xvclNwYWNlCi9EZXZp Y2VHcmF5IC9JbnRlcnBvbGF0ZSB0cnVlIC9CaXRzUGVyQ29tcG9uZW50IDggL0ZpbHRlciAvRmxh dGVEZWNvZGUgPj4Kc3RyZWFtCngB7V1nWyJNEFTJLBIkKxJESYqgRxAk/f9f9XbPRk4Ohdu9Z+p9 mi8GdufmytmZorq7+uJCXoKAIKATApfWS6c5fTMXmvGVetE331yqzds85VCYXqHQFcqsac6hcDQa i0WjYZRZ05zDkVgiaRhGIhYJQ2DNc47GjXQ2l8umjXgUYta0NiLxVCZfqlTLhWwqEaUVos3T9qeJ ENAxI1Oo1putxl0pl4pH9J/15SUBfZ2vNh57/c7DfSlrxGiB/Ol/qMnvadLRRKZ4/zh4Hb8O2rVi GmCB8KST2XKj+zqZTcaDh9ubVCyiO9TWpJu9t/lyMRt1G6VMUv9nkdZ0IlNq9N4+VuvlZPh4l7+O 6w/1FT2IhVpnNF9t1ov3frOcNbSHWm15uerDy+Rzu13NXp9qBQW11jsI73mJdKneG39sdpvl+6BV yWFAnbq5bb9MP7e79Xz0RNsenTB60z06x2mnLtW7DPV2+WvQqub0P2F4VRPUalUz1J17OmG0h1qt 6iJDvd5tPyfPDyhQG7SBPNMGslt/jLt1BbXe56K5gRTvaa9WUL+0+TDXnTeZe3WlNfi13O42H2+9 eimjPW9iqOPpQu2JjkVa1dMhQx3XH+pQ1MgpqOmEWbz1GmUA3nRpMpCn1xlBvZph8CY+YYhWN/vv C4Za8Sb9KaqCOn/3OJyttjsY3mRCTbyaoN4Sb2pWMChqPJW/bQ+nBPXapqgAvClJvIk+wli8iSmq /tue4k1t/jSAw5v2KCoUb7pxeRMSRb3vjE3eRBQVhjdVW88mbyKKCsKbEmmboq4mMBTV5k0mRUXQ m0hgV8oN8ab1breaMm8i7VfvjzAXzJuMbIV401JR1H6jnMXhTfAUFUFvoggMU1TmTVAUVUm//GmA 9SZEaY8Pc5H2Agk2ibQXCKyHBrVjAyLtHULH198R1Pu8iSiq9of5vrRHvAlH2iPeNLB4k0h7vq5k 72Dg0h4ebyqDSnsvStqbW9Fnkfa8D5FP31u8CU7ao0QQFRIlFRVK2vOGRIGiz05IVKQ9nx68r8NY vMmGeirS3leMfPqNR9rjkCiQtJf0SnuUSqb9pwESJDkR9e6RpD2VSgaRtXeIoiJIe2FkaU9l7aFJ e56QqEh7Pm3P+8PY0p6pN2Fl7VFIVLL29v+cvv8Em7UHLO29ctaeSHu+r2XPgG70WbL2PLAE8a1N UQFDoh5prwWStccp+FTtsNpRKhlLezghUcnaC+Lp2xuTPnjtS3swhSXeagesrD2r2kGkvb2V6OcP Iu35iebRsTzSHtXwYEl7XO2gsvYAqx24ShRJ2mv0VGHJEqcgl6W936odNDeqUCqqWbANXu0g0t7R TffcN/830h5KlSgXllCVqFmQm6HaZ72z2UXaO/fJOvk+xZu4INeR9u5QCnI9IVHK2oMJiRLUUpB7 8jo98QZ8aQ+w2oELcpGlPQjLB6WieqLP+huZuNIevpGJ5vY8pmeMnUo2eVGuZAhl5qSi2kYmOAW5 ZGRiQw2ZtcdGJhCFJZ6CXJH2TmRCp13u8CbArD2PtAcSEj0g7WFUO7AXo0h7pz1ap17NKqrXa+8Z JiSacoxM5mOskKibtYdkZMLS3kZ57Ym0d+pT9rPrvxTkwvAmJe0pvYkKclnaQzMyEWnvZ0v0jKtQ qx1gjUysrD1Aac/2KAbK2us6FBXFOFc8is/YxE6/haPP3qw9TI9iRGnvDcZrjz2KLTtoGGnP8Sim TwMgHsWHsvb07+3g8ibUrD1lAIfQRsOC2pH2xKP49PPuh3coqCVr74do/dVlDLUr7ZGRCYy0ZzjS 3geatIftUVzE6O3gehR/ikfxX+0RR292s/ZE2jsKlA9v2iFR+Kw9BCMTWGmP29NhexQjSXvSfsyH je3oEFYKvrQfO4qSL29+qXYAkvbc9mOA0t4KTdpD8ij2SHuO1x6StIfdfgygQ64j7Zm8Can9GBmZ mO3HxKPYlxPw8CBWlSicRzH3Mrc65EJJex6PYsnaO7wkffitFX3Gaz8Wd9qPfcLwpt/aaMBk7VHj eCrIBTMygW4/Jh7FPuxtR4ZgFRW4/RikR/HUbdsqRiZH1ub5b1kU1Wp3D+ZR3HWrHXIGRO2zk7Vn Vjuk8YxMRNo7/2E7fqfltVd7GpkexShtNFTWHnT7MdVGQ6S948vz3Hft6LPrUYwi7WWcagccac/2 KMYyMklmSvW9glyIwxzfyAQnaw9W2hOP4nMPjp/eZ/Em28hEsvZ+Ctzp1x0KiaIU5Ho8ioG89qjd PaKRSR/Mo9gj7a1nVttW8Sg+fYP49o5DIVE4aQ+n2oELciHbj8FLe2Jk8u1ecNYFZvTZzdrr1SEM 4EIRLsh1PYpvgTyKB27WHopHMUMtHsVnPWAn3PRV2isoFfXyhDH++aVmlehe+7EcQmFJWKS9f7NW DlFUJI9isjWkrL0HFM8YJyTKvR3uixwSFY9i/1e6yZuqiFl7tkcxTrWDN2tPPIr9X8zOiHyYG1mP tNfEaaMh0p7zZwzum72sPZH2ggP64uIQbxJpLxDETai5J5abIInDmxCz9lqq/Zh4FAeynK1BaVVH jZxIe0Fi7EDNvKlit22lNho4vEmkvX+yQCLx6zxR1Cn1xKIOudTuXqS9YHB3eBNDvZ6PCOo0Dm9y PIorICn4yB7FIu0F8wjaoyIbmYC2H6NqB1UlClXtYHV6w/ModoxMMLL2kkraQzUy2TJvgvIotoxM +k289mO2kYlk7dmHmY9fHYrq0ZtwKKrltYeUtWdLe3AexR9Oh9wURA0Pt23dk/aioSu901d+75Db vr2BqH22pb3NjgP9dZBAv5m1xyn4S5icClvaW6w3SrmhekvdOyhT8MjWm5arz7mzPnw8wwIYyjaA 645mi8Vs3CV3Ho7CBPAv+TmkColSXnh/PJm+DzuUPwsxaWo4kC7W2v3haNhv3xVUnpCfsAQxFvMm I1uqPXS6nXa9kkvFw9ovD3oUKVk5lSvd1huNWiWfTka13z3ob8fPYiKVLZTK5eJN2ohFND/H1XKj erpwNG6kM9ls5jqJMWeCmmYdiSWSRjIZV3PWfcezsQ6FI1F6RcK0NiDmrLC+CoXCoVDoCmbOPGta I/Tir0HsqwGNydOlV0Cjy7CCgBYI/AdkofBJCmVuZHN0cmVhbQplbmRvYmoKNDIgMCBvYmoKMjM0 NQplbmRvYmoKMzUgMCBvYmoKPDwgL0xlbmd0aCAzNiAwIFIgL1R5cGUgL1hPYmplY3QgL1N1YnR5 cGUgL0ltYWdlIC9XaWR0aCAzMjAgL0hlaWdodCAzMTMgL0NvbG9yU3BhY2UKL0RldmljZUdyYXkg L0ludGVycG9sYXRlIHRydWUgL0JpdHNQZXJDb21wb25lbnQgOCAvRmlsdGVyIC9GbGF0ZURlY29k ZSA+PgpzdHJlYW0KeAHt3VdD47oWBeApEBIgoYZehqENQy+hB/j/v+puSZa0JavszH08ixdkSXMe viPHxtHy/vYNPxCAAAQgAIGqwHf6qU7ChLSAwjM/6XH0lgRI7sePn/Tz4wfWYAkqPUZ6P6em1c8U BNNEpV7im5rudHu9XndGCWIJlrRaY2rxdbqz/cHCoD/X60z9hF/LqNCh+GZ6/cWV4dpwZbFPgFiA Ba54iPimZ2YHy+tbu3u7W2tL891pOoPjWTjOCGi+uYXVzb3D45Ojg+3h4twMLcDMbHRHAg3fcPvg 9Pzq6vzk1+byfBefgJFS9tDwLQ63f59d348ebs6PdlYHdALjCpIl4wMN39rO4d+bx5fX5/vLY/hx oHJb8XXnFhXf7dPb+P3l/uJoe6WP9Vdms6OO7+j87vn98/P9+fbsYGMJ1w8LVP4d8I0/vz5eHi6O doaDXgf3L2U5Par55unkVatP8b2Ork721s3yw/1LRVA9MZjuzi+t7Wq+L813ur9Bdy+4fa7Y0XDE Z1bfKd380Z9veH5Q9bN867tHF3cvzcl7uq/58NfvBHzHF/cvY3fyruDhQdWOnbzreyGfOnmx+qqC 7uT1fA9XJ3TpAF/VLlx9l83J+3AJPgmdmqNWX4duXOjkvXwwn33gk+JpvqlOr7+8sXdCfB906XgB 3z/w7XO+PXz2yQjp5DWrT/G90uob0+o7pj/a5ru4ba4TJvjoiZ/mU3+04Y/eMiHju2pW3/0F+Mpo fhR83uIfWp7v9GpkPvuw+uSODd/K5j7jO9pVlw589tUZHd8vw/c5Vl927Nqvy3HpKBMavsHKpuJ7 +/j6HD/fnYOvjOZHv3+nHUI9zXdtPvsM3yKeNnukfCvgc6tvZ21xDg/r82puhJ4YTNEOITp5/1z7 kxd8DqjccHwHjO8QfGU1N9rwrW4R3+Pb55f6mvxc883gmzanlG2EfHTlfX+6/Xu4o7apgS+r5gY8 39mNWX3gczj1huFboJOX820PF7D66nj0tFlfeSO+3+CT2NEcy0e7I2+e3tWl4+n2DHxCPcU3PUNb mxnfDfikepZP7c21q0/xreKzT0TYrD7Fd9ucvIZvdkZFjET/jf/wJM9He3PVZ9/b483ZwdbqAvgk q8Lxqa3N4JOQ8TmGb3God4YzvgFWH2fKtWO+j7fH6z908oIvJxb0N3yUSzi/pY316rMPfIFQ8UDx mVhHszP8bUR8mytYfUU2O+j47MZ6xfcLfNan8jvFRzvDVyjXgVxgxY6GNR/FOlgu4Rp8dTczg76n jGIdb6Mry4cdQjVGxqc21rtYBzbW1+T0uOVTsQ6WS9gE32R8fmM9nbzYWC/C01ub9WdfEOvAxnqh nou0MT7EOqR4wepDrEPOZmeqSwdiHVZj4t/Eh1jHxGruHzg+xDqcyQSNgA+xjgnk9NQEH2IdckTG h1iHnM3O5HzIJVgV8W/Px3IJx8glCAEbPhfroFzCHeUSEOuQ+Tk+F+tALkEmp2cZPh/r+EKsYwI9 emTAYh08l4BYh4SR/uRtUjFsY/0RNtZL7GiO4tOxDp5LAJ9QL+DTO8PVZx9iHRPyNbEOlUtArENq 509elopBLmFivmhjPaVisDdXhGguHTEfcgkiPHvlNRvrXagIuQShXnPlDXIJiHVI8fTqi2Md4JuY r51LQKRNgkiXDrX6PJ/KJahUDHaGT8aHXILEK5pjV1+cS8Dqi6DSh4YviHVgY32aKtXb8Ol6CSqX gFhHSinb5/l0vQQb60AuISsWDCg+HevguQTEOgKjwkGSz+YSkEctyOmhFJ/dWI9yRTW9dqwDuYSq mZ9A31PyWIfPJahQEWIdHirdsnxxuQnkEtJeUS/nY7EOvPQ6csocOj7EOjJCxe4EH3IJRbFgMMWH chMBUemA8aHcRAkqPUZ8VBq6v+yrdaDcRFoq2av5WtU6UG4iidXudHw21mHLTfS7uG1uc8U9AV8Y 68Ar12Os9nGCDy/8bzPlehgfYh05pHy/5+O5BNSKyYsFI0k+5BICo8JBw4dYR8GoMGT4mlwCXXlR bqKA1R4K+CiXYGMdKDfRtkr0ZGMdKDeR0Gp10bN6xDpaKuIOxYdYh5grnmj5WLmJJtaBegmxVeK4 4QtiHVRugkrtgC/BFXeFfOaN9ajWEStljw1fnEtArCMLFg4k+ZBLCJHyR5ZPvfDfxjpQLyHvFY0Q H3IJkckEhw2f31j//oRyE3K/LB9yCRJEz8dyCYh1SOjUHMeHchNSMj7P8iHWwVXEbcOHWIcYLJzY 8LlyE4h1hD6VI8XHy02Yah2ol1Bhs8OOD7EOSzLJ7yQfYh1SQs1H5SZ22OpDrEOsZ/iWfLUOxDqk djSPvqdMxzr6iHUIGC0fYh0CrPYUz3fMq3VsIBXTtkr0OD7EOhI61a4EH2IdVTU3gfGh3IRTETcU H8pNiLniicSHchMxivzY8dlcwodKxSDWIRQM+MJcQocqa6O0dtkxwYdyE2UyPsr4kEvgMLI2+GRO mVmeD7GODFGpu+FzuYSv8cs9yk2UxIIxx+fKTWi+tSXzwn9ceQOt1oHh8+UmEOtoEZU6glyCvu8z xU4Q6yipubGAj1frQKzDGRUa9MTA5BJ+oVpHgSk35Ph4tQ6Um8hxxf0NH6uXgHITsVHhOORT1TpQ bqLAFQ95Pr8zHLGOWCl7bPjiWAfKTWTBwoEkH2IdIVL+yPKpWMcTvbVZffYh1pH3ikaIL451qFwC Su1ETpnDhk/FOuzqQ6wjY5XoZny3zcmLchMJp0yX52OxjoMtVOvIeEXdjg+xjkhGdGj4glzCzRmt vgEyWRK/mA+5BImam9PwuViHySVg9TmgckPx8VjHx9uIasUg1lFWc6OOj22s/6NyCfjsc0aFRooP uYQCWDik+SjWsatX39fXx+voGnyhUf6IvqfkuQTiQ6wjr9UaYXxqY72v1oFYR8sq0WH5VKyDlZvY RC4hgdXucnzIJbRx6j1JvpN9FDup06kZCT7EOmR0ER9iHXI2O1OtPsQ6rMbEv4kPsY6J1dw/cHyI dTiTCRoBH2IdE8jpqQk+xDrkiIwPsQ45m53J+Ubm5EWxE4tT/+35eKxjdx0b6+t2NKPhc7GOz/HL HWIdIjo1yfG5WIfZWI/VJyI0fD7WYctNgE/IZ+slnF6PeC7BhIpE/43/8CT6kxflJv79/7/iQ7mJ f/bjfPqd4WO6dCDWIfVs+IJyE4YP5SYEhiEfyk0IyPgUwxfnElBughsV2mk+7AwvkPEhy8fLTfzF xnpOVGpzPsQ6SlLJMeKLYx1IxSSlkp0NH6p1JHWqnS2+t0fEOqpqboLnQ6zDocgblg/lJuRmbKbh C2IdtDMcG+sZUanZ8FEu4e/tM+VREesoabXGPN/5neJrYh3YWN+SSnYoPh3r4LkExDqSVonOJB/K TSSkkl0pPuQSklSpTs3HYx3IJaSYMn30PSWPdfhcwgDlJjJkvNvyodwEVxG3OR+LdWBjvUzQ8SHW IQMLZyX4kEsIiUpHKb5LpGJKZHyM8V0+vIxVHpVe+I9QETcqtIlvqtPtL6/vHTd8ql4C+ApkfEjz 9frLG3snxPdBqw/lJrhPpe34bKxjTHzHe+tL/W5nCuUmKnp2c+nyhuILYx3TqNZR00vxYWN9Vc1N YCcvYh1ORdzwfDyXQJ99eOG/xDDJd4RYh8SO5jR8iHUIvaJphq/JJdCVF+UmIqDyYcxnYx0oN1F2 a0ZRbkLElJtEz+oR68jh1PsVH2IddafMDMu3RdU6dKzj3cY6kEvIkPHuho/FOnS5ibXFOfBxp0w7 5EOsI8OU6zZ8iHXkfCr9ST7kEipqbtjy8VgH6iU4nlqD+JBLqCHlxxs+5BLyRKWRLB/e2lxis2Oe j+USfm+j3IQFKv92fCg3UYZKjxq+IJegy00s4ORNg4W9nI9trEe1jpApd9TwuXITiHXkpJL9io+X mzCxDtRLSGK1Ox0fYh1tnHpPkg+xjjqcmaH5KJeww1YfYh1iPcO35Kt1INYhtaN59D1lOtaBchMS RcuHWIdEqzXH8x3zah2IdbSkkh2Oj8c6sLU5aZXoTPAh1pFwynQxPpSbyBgVuhUfyk0UgMpDxIdy E2Wi0qjjs7kExDpKXPFYwBfmEjrIJcRareMEH8pNtJSyHYwPuYSsUnYAfFkayYDnQ6xD4hXNafhc LuFr/HKPchMRUv7Q8blyE5pvDaGivBkbMXy+3ARiHQyn3gxyCfq+zxQ7QayjbkczPN+fsFrHXFfF eUX/jf/wJHpiYHIJv8D3D8vA8elcQhNpQ7kJqWTD1+QS1Mb651uUm5DqfWvz6VjHELEOEaHha+cS wPd/8KHchAiPblz0lXdhdfvg7MZE2ujkRaxDqOf5fp/doNyEVM3No9UXxzrUC/+HC0gEOqNCo+FD rKNgVBjK8mFneEHNDXk+FuugF/5jY70jKjXokYH67BtuI9ZRYsqNqed9mu93uPoGOHlzZEG/2mUw S/d9DR9yCYFO9UCdvd3+yhbdNuv7vrdHFDuporEJ+uxdWt8/1S/S+HgbER9iHQyo3DTLb3X7UO0u /fx4JT6VS8BnX5nNjSq/3sLa3snV6HX8TnzIJTgbSYPu/Tqzi+r0HT2/PN9fgU+i5udoP1p/xxe3 D/e3Fyf7myvIJXieakufvwO6ezm9uLw4O9rDxvoqWTCB/Ka680vrOwdHx4cHO+vLWH2BT/VAncD0 8vC1rZ2d7Y3VxXnUiqmSBRPUApyZ7S+uDIcrS4M5vDM80BEc6Kcvvbn+YNCf6+n3p2GXgYDNTaHn B/QqyZlur9ed6Uxha7ODkTYIkASn6Yf08MJ/KZuf910L/iQ86HmVSVok2PxM8q8wlwmQHztCEwIQ gAAEIJAV+B8Y+7V6CmVuZHN0cmVhbQplbmRvYmoKMzYgMCBvYmoKMzY0OAplbmRvYmoKNDUgMCBv YmoKPDwgL0xlbmd0aCA0NiAwIFIgL1R5cGUgL1hPYmplY3QgL1N1YnR5cGUgL0ltYWdlIC9XaWR0 aCAyOTYgL0hlaWdodCAyOCAvQ29sb3JTcGFjZQovRGV2aWNlR3JheSAvSW50ZXJwb2xhdGUgdHJ1 ZSAvQml0c1BlckNvbXBvbmVudCA4IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4Ae2Y SXOCQBCF2YdlQMBBDSCGMlEsjQtUmUgSk///qzIzCOKSswf6XTzA6eN1+14LAggIAIFHEhBBFwTu fgv+hgQ6E2BEblCJoiTJsqyAagKyLEnXpDglRdUQQjqIEqAgNFXhpNqWEkVZUZFhWtgGcQIYmwbS KKn27FE7KZpuOa7fJyQABQHp+66DDaRSUI2hKCZZM7BLhmE8TpJJ55Uk4zgcBZ5tIqVlKDp1qo7d QTSZvs6zBWiRzWcvz/HQt42WofjUmT0SpbPlersrQEW+26yyaTxwrZahGCeEvdFktsr3H4fys/Mq y8O+WGdp2Ld1tRk8yklF2H9KF9t9+XX8Af0cv8v3fDmNiKOrcv2Pd+IU1px+O6+aU0wcuqBanE5z 98bmrvNTRwH8N3eKZvaCKJ0vN7sc1nhx2uPjyz0uVLnAg1zQJKIsq3OB1gqaPBg0ObPzKZMCSJI4 us2Z1FC0txhVb4HWwggQ4nu9694iCLS48B6MWQ92Oi/WhLFl6Nc9mBmKkaJ3FbiqVASqu8rluYAV Yk4KznT1kY7+3rvTcVAMFehM4O7dt/IUewSqCTAmICAABB5H4A8dKHb0CmVuZHN0cmVhbQplbmRv YmoKNDYgMCBvYmoKNDQ1CmVuZG9iago0MyAwIG9iago8PCAvTGVuZ3RoIDQ0IDAgUiAvVHlwZSAv WE9iamVjdCAvU3VidHlwZSAvSW1hZ2UgL1dpZHRoIDI5NiAvSGVpZ2h0IDI4IC9Db2xvclNwYWNl Ci9EZXZpY2VHcmF5IC9JbnRlcnBvbGF0ZSB0cnVlIC9CaXRzUGVyQ29tcG9uZW50IDggL0ZpbHRl ciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngB7ZjbeqIwFEaVU8IhQY6KYKGMlaK1BdQ6oK19/6ea JG0F/HrduSDrAgG5WvzZ7OzRiMMNcAP/0cCYc2Pgp5dBHxE4XQPEyK0pJkmUCDKHGiAmRFEQbkzR KEmyAgCEUOWoKoQAKLIk3ogaC6KkQFU3EMYmx8QYIV2DitwXxTSphmk5nuf708Hj+77n2ibSABV1 rVFk1UmKiixvFi7iOOEkSXwXBb6NNSCJbYkaj0UZGpYfJcssz9ebwbNe59lDupjZWFXaQLE4aaYX pdlTUe32h8Gz31XFNl8uppYOpGspp56AYc2SbLt7rZvTefCcTvVx/5Lfzx0E5evCI55kYNjzNC9e m/P75WPwXN7fTsdqs4xcrPY8kTzZwX1eHOrz24VDPDXE05/Qxf08SUC3pvHqqTwc64bT1H8P1fNj EtB116tPiobdebJab1/Ksho8ZVk8b7I08ifdOj767Asm3vwuXa6y7HHwZNnqIY1D30KkPLWNJink okLacccPwmjBIQaiMJi6E6QqbVtA2nLWQUEdmZbtuBxiwLGtCdaJpk6cvkTJAGq6YRiIg4gGXVMB 09Tu72ig6FZYppMVzqcBhY1Vrh+7780wESUIoshGdXRcN3iICzqm66aJuSK3mKvu4HPQ50zId4z6 v+wvfvgy0HfDr7gBbuDXDfwDXdN7GgplbmRzdHJlYW0KZW5kb2JqCjQ0IDAgb2JqCjUxMAplbmRv YmoKMzMgMCBvYmoKPDwgL0xlbmd0aCAzNCAwIFIgL1R5cGUgL1hPYmplY3QgL1N1YnR5cGUgL0lt YWdlIC9XaWR0aCAxNzQgL0hlaWdodCAzMzAgL0NvbG9yU3BhY2UKL0RldmljZUdyYXkgL0ludGVy cG9sYXRlIHRydWUgL0JpdHNQZXJDb21wb25lbnQgOCAvRmlsdGVyIC9GbGF0ZURlY29kZSA+Pgpz dHJlYW0KeAHtXdlWIlkQbPZ9R1ZZFFFREYFmR/T/v2oyb+2FdrdFPUSeuTzMEXqqzj1RyV2CiMhf v/RLI6AR+L8iELFeEgCgsUbVi/5AHy+PNRbnVywahR8ujzWRTKVSyQSNF3y4PNhkOpvL53OZVAJ8 uJFILJ7K5kuVaqVcyKV5uLilSzUbT2aL1Uar3WrUSjlCF7gWqA6SmWKt1RveDHvtWjGbRAaXoE3n q63+3ePk8a7fquRTcdxSYGiz5UZ//Dx7e53cdmsFqlzYwo3EEulCrTt6XqzWy+m4d1XM4I7WgLY5 nMw3+91qdk+jBcY2EiVo69fj1/Xh/bCe3V/XC2nYujWgbQ2fFrvT6bCa3nWqwN8ynhAKNYJ2c/w4 7ZdPw1YZdwazqvZpsT99vm+pbKkQEqgTmAFtl6r2/fPj8Pv5tl3JwU63FrSTxf7j87R9e+gb8xfm dKuWsVr37nV9JGhXL6NOFRzaUnNA0J4I2vlDv6GWBlhoU3mCdsrQHtGhjdAOodQYPM53BO1u/thv lHjVBYa22rmbro6fn8fVlKoWd2WgI0Mi44Z2IADaEUH7wdBCL7omtH2zahePBC0tusBVm6t2Ri8K 2vX0rlvLw+6+FLTFq/7DfEsTwn4xGTTRoa20CdoDVe36VQa0bya0wyb0VjGeylXaty+/Cdr39ev4 mk6PqGcGs2p7D29b2iruF08ioH1W0G4IWujjGB3Li1e9ewPa5dMN+gEnV2ndmNDO8KEt1Hv3sw1X rQBos+XWzdOSJ4SNQSIAnx0NxkNBS2fHm1YlB0ss2ozHks6ODG2vDswmqWM5k0lcteaxPAm7MigK tElkknEsZ6IOl1a0ySSRjIcJLezZMc3HcovxwCeTSsQu24wHOJnkYzyQeTqT8ZgQu2wyHvBkEjEe RCaZjAcwT0fQimQ8FLSSGI+5YjzQeTpBjEepQWQSH8t3EhiPaudWMx6h6y9o8+UwHkQmjfn3cth9 LbPLdCwnxuPEjMdEGOMBDO0vE1qb8YD/SZd4OkGMR5EZD8XTSWA8iEwSxXgYZNJhKYLxIGhtxgNY m2QxHhaZxGdHfMZDQbt9w5Ym2YyHRDLpAZ7xID0dMR6Kp3vBZzzKNuMhTj4Dz3hY8hml8YBnPGz5 DGs8BDEeEuQzrPEQw3g4yiQR8hm3xgNXqmgok2z5jACNhzj5jFfjgaxMYo2HI5+BZzz6mvEIn/kS xnjEk6xMkqPxYPmMGMZDyWdkaDzYMOTIZ4jxKMNrPHyMB671wiufuWX5DCwXrjQeZBiSw3hIlc9Y jEfoS1AoNzTIJG0YCgVM700YWrd8ZgTvxbINQzsyDMmSz4AzHtow5P1uhPROG4ZCAvL8NtowdI5J SJ9ow1BIQJ7fxobWSyaBnhloh+DIZ9gwJEg+Q7Z9CV4sxzAELp/RhqHzb3MYnxBP52Y88L1YHsMQ eNiEl/EQZhhCjp9Rx3LNeIQxAfjvYTAeHOzjeLFwf3c0vVgyI1Lw5TOCDUMCGA9BhiGOSHHSZ6Aj UmQahlREioT0GUvjIdAwhLuMmfKZB5WZJCIiRclndESKf/d04XsP4yHIMMRhE9owdOHDd18uTD4T Y/mMNMOQmIgUmYahvYyIFLd8RkBEijzDECWXColIcRmGcLeKvvQZSz6DG5HiGIZWAgxDkhgPr3ym I8YwtOVQWDmGIfAoY9ERKRIYD20Ycp9MwvnbL5/RhqFwcKW7fKPx0IahyxE2odURKZdD6b+DBa2U iBRxhiFJESnSDEPXVt6uRMMQavsmq8OQzXjIydsVEZEiXT4DzXho+Yx/gb/8PSuT5BiGVCisE5Gi DUOXF4B5B9oqGh2GnLxd/A5DdkSKZjzCKwTyB+TtHA/qMNTFb95kyWdkhMIKytuV1mHIHZEioHmT qFBYu8OQNgyFNdnS5ssdkYLdvInzdrVhKKxH772PbMMQ98XSHYa8TzTYO7PDkDD5jKwOQ5w+w41/ dYehYDV6fpXsnsraMHT+RAN+YpFJLlcLdihs1pU+I6GnspG3Sx2GWD4DbRiS2VPZat4ED63FeBiG Ieieyg6ZZPRUxlWBKn9Ag3oqc8sxbRgKuAh8ddk38hkJESmCDEOqw5COSPmqAIN95hiGGFpKn8Hv qawNQ8Ee9R+vUgccbRj6I0YB/1Em42HLZ8R0GNKGoYAF+uVlFuNhtau2yCRo+YzZYUiCYcjoMMQ8 HX4orFc+ow1DX35hAnzItn1WJpk9lbVhKACG31yioCWNh9VhSJNJ3wD1848taHWHoZ9j95crGNqc rUwyyCT0xr9EJnG7apXjUcomY+iMh+4w9Jci/PE/q6q12+BQ3u41eCgst6smw5DJeIDn7frkMyRE QO1+YshntGHox1+gf7jAdLW4DUPIcy0prcvcrlpGT2Vf3i4zHrjLmILW6DBkMh7pBK4XK04p0RQK 64pIAW/e1ByKYjwsw5CAiBSH8ZDWUxm6w5A2DP3D8hnof7EMQ668XW0YCoTk2UXq1/JSQ15EitGu WhuGzp5owA9kMx6DJjMeciJStGEoYJn6LrMYD20Y8gFz+VuC1sd4gJNJpmFIy2cuf/aeOwg1DJmM hwTDkEEmHSTl7Z62s/teXUDerizGQ5BhSPdU9syS4byhRdej8dCGoXBwpbtY0AozDLE/YMd5u/CM B8lnVsfPD20YCq1kjapVESmS5DOCDEN+jYc2DIVSvMaxvEPpM7rDUCiAOjcxoG3b6TOSeiovtGHI eZCX/iXMMOQhk2ZjPpbjy2ekdBgSlLcbVRoPk/FYUt4uvsbDLZ8RwHgIMgy1PPKZDLZ8puaVz+Bu FSV2GCL5DJ0dD3IMQ3Qsp2Cf/hV2GxwPmTSSYxjakRdLdxi6dItoXn+m8YAnk6gZpTo7Epk0hTYM EavI+5ke72c+mKcbIPN0rAvPkNiHdFTvpwM+tKZ4ebE9HjZvDC2uxoPpZfq5nKTW8/V2s3jGrlpr tMPH18Vy/jLuKWhBDUNM3XNg+FVvNHmZPt8PWhXczCSacnm6TReqrf5oPB7127UC7WthoVW/i6Sy pVqr2+t1m7ViFlYWrlYzKoVEOles1htX9Uoxm4qjir6M0fJwU9l8sVQq5jMp6DpQlUvDTaYz2Uwm lYxT0WJmDChozeHGE4lkMkFjRR8szwuRaDRGLx4rNrJm8dKAaaQSxsoDVkMVgKsCV/9HIyAGgf8A FtkNhAplbmRzdHJlYW0KZW5kb2JqCjM0IDAgb2JqCjIzOTIKZW5kb2JqCjQ3IDAgb2JqCjw8IC9M ZW5ndGggNDggMCBSIC9OIDMgL0FsdGVybmF0ZSAvRGV2aWNlUkdCIC9GaWx0ZXIgL0ZsYXRlRGVj b2RlID4+CnN0cmVhbQp4AZ2Wd1RT2RaHz703vdASIiAl9Bp6CSDSO0gVBFGJSYBQAoaEJnZEBUYU ESlWZFTAAUeHImNFFAuDgmLXCfIQUMbBUURF5d2MawnvrTXz3pr9x1nf2ee319ln733XugBQ/IIE wnRYAYA0oVgU7uvBXBITy8T3AhgQAQ5YAcDhZmYER/hEAtT8vT2ZmahIxrP27i6AZLvbLL9QJnPW /3+RIjdDJAYACkXVNjx+JhflApRTs8UZMv8EyvSVKTKGMTIWoQmirCLjxK9s9qfmK7vJmJcm5KEa Wc4ZvDSejLtQ3pol4aOMBKFcmCXgZ6N8B2W9VEmaAOX3KNPT+JxMADAUmV/M5yahbIkyRRQZ7ony AgAIlMQ5vHIOi/k5aJ4AeKZn5IoEiUliphHXmGnl6Mhm+vGzU/liMSuUw03hiHhMz/S0DI4wF4Cv b5ZFASVZbZloke2tHO3tWdbmaPm/2d8eflP9Pch6+1XxJuzPnkGMnlnfbOysL70WAPYkWpsds76V VQC0bQZA5eGsT+8gAPIFALTenPMehmxeksTiDCcLi+zsbHMBn2suK+g3+5+Cb8q/hjn3mcvu+1Y7 phc/gSNJFTNlReWmp6ZLRMzMDA6Xz2T99xD/48A5ac3Jwyycn8AX8YXoVVHolAmEiWi7hTyBWJAu ZAqEf9Xhfxg2JwcZfp1rFGh1XwB9hTlQuEkHyG89AEMjAyRuP3oCfetbEDEKyL68aK2Rr3OPMnr+ 5/ofC1yKbuFMQSJT5vYMj2RyJaIsGaPfhGzBAhKQB3SgCjSBLjACLGANHIAzcAPeIACEgEgQA5YD LkgCaUAEskE+2AAKQTHYAXaDanAA1IF60AROgjZwBlwEV8ANcAsMgEdACobBSzAB3oFpCILwEBWi QaqQFqQPmULWEBtaCHlDQVA4FAPFQ4mQEJJA+dAmqBgqg6qhQ1A99CN0GroIXYP6oAfQIDQG/QF9 hBGYAtNhDdgAtoDZsDscCEfCy+BEeBWcBxfA2+FKuBY+DrfCF+Eb8AAshV/CkwhAyAgD0UZYCBvx REKQWCQBESFrkSKkAqlFmpAOpBu5jUiRceQDBoehYZgYFsYZ44dZjOFiVmHWYkow1ZhjmFZMF+Y2 ZhAzgfmCpWLVsaZYJ6w/dgk2EZuNLcRWYI9gW7CXsQPYYew7HA7HwBniHHB+uBhcMm41rgS3D9eM u4Drww3hJvF4vCreFO+CD8Fz8GJ8Ib4Kfxx/Ht+PH8a/J5AJWgRrgg8hliAkbCRUEBoI5wj9hBHC NFGBqE90IoYQecRcYimxjthBvEkcJk6TFEmGJBdSJCmZtIFUSWoiXSY9Jr0hk8k6ZEdyGFlAXk+u JJ8gXyUPkj9QlCgmFE9KHEVC2U45SrlAeUB5Q6VSDahu1FiqmLqdWk+9RH1KfS9HkzOX85fjya2T q5FrleuXeyVPlNeXd5dfLp8nXyF/Sv6m/LgCUcFAwVOBo7BWoUbhtMI9hUlFmqKVYohimmKJYoPi NcVRJbySgZK3Ek+pQOmw0iWlIRpC06V50ri0TbQ62mXaMB1HN6T705PpxfQf6L30CWUlZVvlKOUc 5Rrls8pSBsIwYPgzUhmljJOMu4yP8zTmuc/jz9s2r2le/7wplfkqbip8lSKVZpUBlY+qTFVv1RTV naptqk/UMGomamFq2Wr71S6rjc+nz3eez51fNP/k/IfqsLqJerj6avXD6j3qkxqaGr4aGRpVGpc0 xjUZmm6ayZrlmuc0x7RoWgu1BFrlWue1XjCVme7MVGYls4s5oa2u7act0T6k3as9rWOos1hno06z zhNdki5bN0G3XLdTd0JPSy9YL1+vUe+hPlGfrZ+kv0e/W3/KwNAg2mCLQZvBqKGKob9hnmGj4WMj qpGr0SqjWqM7xjhjtnGK8T7jWyawiZ1JkkmNyU1T2NTeVGC6z7TPDGvmaCY0qzW7x6Kw3FlZrEbW oDnDPMh8o3mb+SsLPYtYi50W3RZfLO0sUy3rLB9ZKVkFWG206rD6w9rEmmtdY33HhmrjY7POpt3m ta2pLd92v+19O5pdsN0Wu067z/YO9iL7JvsxBz2HeIe9DvfYdHYou4R91RHr6OG4zvGM4wcneyex 00mn351ZzinODc6jCwwX8BfULRhy0XHhuBxykS5kLoxfeHCh1FXbleNa6/rMTdeN53bEbcTd2D3Z /bj7Kw9LD5FHi8eUp5PnGs8LXoiXr1eRV6+3kvdi72rvpz46Pok+jT4Tvna+q30v+GH9Av12+t3z 1/Dn+tf7TwQ4BKwJ6AqkBEYEVgc+CzIJEgV1BMPBAcG7gh8v0l8kXNQWAkL8Q3aFPAk1DF0V+nMY Liw0rCbsebhVeH54dwQtYkVEQ8S7SI/I0shHi40WSxZ3RslHxUXVR01Fe0WXRUuXWCxZs+RGjFqM IKY9Fh8bFXskdnKp99LdS4fj7OIK4+4uM1yWs+zacrXlqcvPrpBfwVlxKh4bHx3fEP+JE8Kp5Uyu 9F+5d+UE15O7h/uS58Yr543xXfhl/JEEl4SyhNFEl8RdiWNJrkkVSeMCT0G14HWyX/KB5KmUkJSj KTOp0anNaYS0+LTTQiVhirArXTM9J70vwzSjMEO6ymnV7lUTokDRkUwoc1lmu5iO/kz1SIwkmyWD WQuzarLeZ0dln8pRzBHm9OSa5G7LHcnzyft+NWY1d3Vnvnb+hvzBNe5rDq2F1q5c27lOd13BuuH1 vuuPbSBtSNnwy0bLjWUb326K3tRRoFGwvmBos+/mxkK5QlHhvS3OWw5sxWwVbO3dZrOtatuXIl7R 9WLL4oriTyXckuvfWX1X+d3M9oTtvaX2pft34HYId9zd6brzWJliWV7Z0K7gXa3lzPKi8re7V+y+ VmFbcWAPaY9kj7QyqLK9Sq9qR9Wn6qTqgRqPmua96nu37Z3ax9vXv99tf9MBjQPFBz4eFBy8f8j3 UGutQW3FYdzhrMPP66Lqur9nf19/RO1I8ZHPR4VHpcfCj3XVO9TXN6g3lDbCjZLGseNxx2/94PVD exOr6VAzo7n4BDghOfHix/gf754MPNl5in2q6Sf9n/a20FqKWqHW3NaJtqQ2aXtMe9/pgNOdHc4d LT+b/3z0jPaZmrPKZ0vPkc4VnJs5n3d+8kLGhfGLiReHOld0Prq05NKdrrCu3suBl69e8blyqdu9 +/xVl6tnrjldO32dfb3thv2N1h67npZf7H5p6bXvbb3pcLP9luOtjr4Ffef6Xfsv3va6feWO/50b A4sG+u4uvnv/Xtw96X3e/dEHqQ9eP8x6OP1o/WPs46InCk8qnqo/rf3V+Ndmqb307KDXYM+ziGeP hrhDL/+V+a9PwwXPqc8rRrRG6ketR8+M+YzderH0xfDLjJfT44W/Kf6295XRq59+d/u9Z2LJxPBr 0euZP0reqL45+tb2bedk6OTTd2nvpqeK3qu+P/aB/aH7Y/THkensT/hPlZ+NP3d8CfzyeCZtZubf 94Tz+wplbmRzdHJlYW0KZW5kb2JqCjQ4IDAgb2JqCjI2MTIKZW5kb2JqCjggMCBvYmoKWyAvSUND QmFzZWQgNDcgMCBSIF0KZW5kb2JqCjQ5IDAgb2JqCjw8IC9MZW5ndGggNTAgMCBSIC9OIDMgL0Fs dGVybmF0ZSAvRGV2aWNlUkdCIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4AYVV32/b VBQ+iW9SpBY/IFhHh4rFr1VTW7kbGq3GBkmTpe1KFqXp2Cok5Do3iakbB9vptqpPe4E3BvwBQNkD D0g8IQ0GYnvZ9sC0SVOHKqpJSHvoxA8hJu0FVeG7dmInU8Rc9frLOd855zvnXttEPV9ptZoZVYiW q66dzySVk6cWlJ5NitKz1EsD1KvpTi2Ry80SLsEV987r4R2KCMvtke7+TvYjv3qL3NGJIk/AbhUd fRn4DFHM1Gu2SxS/B/v4abcG3PMc8NM2BAKrApd9nBJ40ccnPU4hPwmO0CrrFa0IvAY8vNhmL7dh XwMYyJPhVW4buiJmkbOtkmFyz+Evj3G3Mf8PLpt19Oxdg1j7nKW5Y7gPid4r9lS+iT/XtfQc8Euw X6+5SWF/BfiP+tJ8AngfUfSpkn103udHX1+tFN4G3gV70XCnC037anUxexwYsdH1JeuYyCM413Vn ErOkF4DvVvi02GPokajIU2ngYeDBSn2qmV+acVbmhN3Ls1qZzAIjj2S/p83kgAeAP7StvKgFzdI6 NzOiFvJLV2turqlB2q6aWVEL/TKZO16PyCltu5XClB/LDrp2oRnLFkrG0ekmf61memcR2tgFu54X 2pCf3dLsdAYYedg/vDov5gYc213UUmK2o8BH6EREI04WLWLVqUo7pFCeMpTEvUY2PCUyyISFw8th MSJP0hJs3Xk5j+PHhIyyF70tolGlO8evcL/JsVg/U9kB/B9is+wwG2cTpLA32JvsCEvBOsEOBQpy bToVKtN9KPXzvE91VBY6TlDy/EB9KIhRztnvGvrNj/6GmrBLK/QjT9AxNFvtEyAHE2h1N9I+p2tr P+wOPMoGu/jO7b5ra3T8cfON3Yttxzawbsa2wvjYr7Et/G1SAjtgeoqWocrwdsIJeCMdPVwB0yUN 62/gWdDaUtqxo6Xq+YHQIybBP8g+zNK54dCq/qL+qW6oX6gX1N87aoQZO6YkfSp9K/0ofSd9L/1M inRZuiL9JF2VvpEuBTm7772fJdh7r19hE92KXWjVa581J1NOynvkF+WU/Lz8sjwbsBS5Xx6Tp+S9 8OwJ9s0M/R29GHQKs2pNtXst8QQYNA8lBp0G18ZUxYSrdBZZ25+TplI2yMbY9COndlyc5ZaKeDqe iidIie+LT8TH4jMCt568+F74JrCmA/X+kxMwOjrgbSxMJcgz4p06cVZF9Ap0m9DNXX4G3w6iSat2 1jbKFVfZr6qvKQl8yrgyXdVHhxXNNBXP5Sg2d7i9woujJL6DIo7oQd77vkV23Qxt7ltEh//CO+tW aFuoE33tEPW/GtqG8E585jOiiwf1ur3i56NI5AaRUzqw3/sd6Uvi3XS30XiA91XPJ0Q7Hzca/643 GjtfIv8W0WXzP1kAcXgKZW5kc3RyZWFtCmVuZG9iago1MCAwIG9iagoxMDQ3CmVuZG9iago3IDAg b2JqClsgL0lDQ0Jhc2VkIDQ5IDAgUiBdCmVuZG9iagozIDAgb2JqCjw8IC9UeXBlIC9QYWdlcyAv TWVkaWFCb3ggWzAgMCA3OTIgNjEyXSAvQ291bnQgMSAvS2lkcyBbIDIgMCBSIF0gPj4KZW5kb2Jq CjUxIDAgb2JqCjw8IC9UeXBlIC9DYXRhbG9nIC9QYWdlcyAzIDAgUiAvVmVyc2lvbiAvMS40ID4+ CmVuZG9iagoxMCAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VG b250IC9LVVZVTkQrQ2FsaWJyaSAvRm9udERlc2NyaXB0b3IKNTIgMCBSIC9Ub1VuaWNvZGUgNTMg MCBSIC9GaXJzdENoYXIgMzMgL0xhc3RDaGFyIDQyIC9XaWR0aHMgWyA1NzkgMjI2IDQ1OQo1MzMg NjE1IDU0NCA0ODggNDY4IDQ4NyA1MTkgXSA+PgplbmRvYmoKNTMgMCBvYmoKPDwgL0xlbmd0aCA1 NCAwIFIgL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBXZHLasMwEEX3+got00Ww7Dya gBGUlIAXfVC3HyBL41RQy0JWFv77XikmhS7uwJmZOx6PilPz3DgbefEeRt1S5L11JtA0XoMm3tHF OlZW3FgdF8o5PSjPCpjbeYo0NK4feV0zzosPWKYYZr56MmNHDyn3FgwF6y589XVqc6a9ev9DA7nI BZOSG+ox7kX5VzUQL7J13RjUbZzXcP11fM6eODaCo7ytpEdDk1eagnIXYrUQsj6fJSNn/pUWQ9fr bxVYXVUSzeLIEQxCJRCUyM6l53D7RNcvs6tS1klCbEuJCRsgBNwn3AJ3GTcJ90AI1SrhIxAC7hIe gJAQO5XwCISAx4QKCAEPy0K3FdJfpevfr6WvIeBQ+YnyDdNtrKP7K/rRp1tk/QIRLpO9CmVuZHN0 cmVhbQplbmRvYmoKNTQgMCBvYmoKMjk5CmVuZG9iago1MiAwIG9iago8PCAvVHlwZSAvRm9udERl c2NyaXB0b3IgL0ZvbnROYW1lIC9LVVZVTkQrQ2FsaWJyaSAvRmxhZ3MgNCAvRm9udEJCb3ggWy01 MDMgLTMwNyAxMjQwIDk2NF0KL0l0YWxpY0FuZ2xlIDAgL0FzY2VudCA5NTIgL0Rlc2NlbnQgLTI2 OSAvQ2FwSGVpZ2h0IDYzMiAvU3RlbVYgMCAvWEhlaWdodAo0NjQgL0F2Z1dpZHRoIDUyMSAvTWF4 V2lkdGggMTMyOCAvRm9udEZpbGUyIDU1IDAgUiA+PgplbmRvYmoKNTUgMCBvYmoKPDwgL0xlbmd0 aCA1NiAwIFIgL0xlbmd0aDEgMTU4NDAgL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngB 1Zt5QFT13v+/ZxZmYGZgBphhGWAGjgwiICouuCSjLIq4sYwBbqyKuyK4o6SVRZm272X7QuVh0MJW K9sX2251W+3e6tYty/auBfze53z4uPR7ep4/nn96Bt7zfn0+32W+2zmDI7Q0tzYKi2gXejGsfkXt aqE9cl6GZdava/FSnNokhDFh0erFKygecq4QofGLl29cRPFIO2KlqbG2gWLxO3x0ExIUSyPhg5pW tGygOGcvPHP5qvqB8pxfEceuqN0w8PriA8TelbUrGuF4nHUIT97VzY0D5VIlurO3bGhs0aFgbeuq FeySGjesqJdEuBBqYJG8wi6eFiahg2cL9OjYI20XBpSq5cabenb98pJjYcSEn0ScGQkhHvlqizp/ 8dzOuy/67UTvRaFfmx5EGIoe6IF2ppt63xMibO9vJ07sDf1a62mgUDNLMFTv7dGduz80VpoG2MGw neEchnaGbQxbGdoYtjBsZtjEsJFhA8N6hnUMrQwtDGsZ1jCsZljFsJJhBcNyhmUMSxmWMDQxLGZY xNDI0MBQz1DHUMtQw7CQYQHDfIZ5DHMZqhmqGCoZzmaYwxBgqGAoZyhjKGWYzTCLYSbDDIbpDCUM 0xiKGaYyTGEoYihkKGDIZ5jMMInBz5DHMJHhLIYJDOMZxjGMZchlGMMwmmEUw0iGHIYRDMMZhjFk MwxlyGLIZMhgGMKQzjCYIY3Bx5DKMIhBZkhhSGbwMngYkhgSGRIY3AzxDHEMsQwxDC4GJ0M0QxRD JIODwc4QwRDOYGOwMlgYwhhCGcwMJoYQBiODgUHPoGOQGMQASP0MfQy9DL8z/MZwguE/DL8y/MLw M8NPDD8y/MDwPcN3DMcZvmX4huEYw9cMXzH8m+FLhi8Y/sXwOcNnDJ8y/JPhHwyfMBxl+JjhI4YP GT5geJ/hPYa/M7zL8A7D2wx/Y3iL4U2GNxheZ3iN4QjDqwyvMLzM8BLDiwwvMDzP8BzDswzPMBxm eJrhKYYnGQ4xPMHwOMNjDI8yPMLwMMNBhh6GhxgeZDjAsJ+hmyHI0MWgMOxjeIDhfob7GDoZ7mW4 h+FuhrsY7mS4g+F2htsYbmW4hWEvw80MNzHcyHADw/UM1zFcy3ANw9UMVzFcyXAFw+UMlzFcyrCH YTfDJQy7GC5muIihg+FChgsYdjKcz3Aew7kMOxi2M5zD0M6wjWErQxvDFobNDJsYNjJsYFjPsI6h laGFYS1DM8MahtUMqxhWMqxgWM6wjGEpwxKGJobFDIsYGhkaGOoZ6hhqGWoYFjIsYJjPMI9hLkM1 QxVDJcPZDHMYAgwVDOUMZQyzGWYxzGSYzlDCMI2hmGEqwxSGIoZChgKG/G71p2X81BxMmujBz8zB JCdsO0XnBJPGIWqnaBvZ1mCSFck2iraQbSbbRLYxmDgJVTYEE/Nh68nWkbVSWQtFa8maKbkmmDgZ DVaTrSJbSVVWkC0nWxZMKETNpWRLyJrIFpMtCiYUoEojRQ1k9WR1ZLVkNWQLyRZQu/kUzSObS1ZN VkVWSXY22RyyAFkFWTlZGVkp2WyyWWQzyWaQTScrIZsWdBdjDsVkU4PuaYimkBUF3SWICoPu6bAC snyyyVQ2idr5yfKo3USys8gmUM3xZOOo+ViyXLIxZKPJRlFnI8lyqJcRZMPJhlFn2WRDqV0WWSZZ BtkQsnSywWRp1LWPLJX6HEQmk6VQ18lkXmrnIUsiSyRLIHOTxQfjZ2Kx4shig/GzEMWQuSjpJIum ZBRZJJmDyuxkEZQMJ7ORWanMQhZGFkplZjITWUgwbjZe3RiMK4UZyPSU1FEkkQnNpH6yPq2K1EvR 72S/kZ2gsv9Q9CvZL2Q/k/0UjK3w9Eg/BmPLYT9Q9D3Zd2THqexbir4hO0b2NZV9RfZvSn5J9gXZ v8g+pyqfUfQpRf+k6B9kn5AdpbKPyT6i5IdkH5C9T/YeVfk7Re+SvROMORtTeTsYMwf2N7K3KPkm 2Rtkr5O9RlWOkL1KyVfIXiZ7iexFqvIC2fOUfI7sWbJnyA6TPU01n6LoSbJDZE9Q2eNkj1HyUbJH yB4mO0jWQzUfouhBsgNk+8m6g648TDoYdM2FdZEpZPvIHiC7n+w+sk6ye4Mu3PWle6iXu8nuorI7 ye4gu53sNrJbyW4h20t2M3V2E/VyI9kNVHY92XVk15JdQw2upugqsivJrqCyy6mXy8gupbI9ZLvJ LiHbRXYx1byIog6yC8kuINtJdn7QWYu5nxd01sHOJdsRdC5CtJ3snKAzgKg96MSbjbQt6BwN20rW Rs23ULvNZJuCzgZU2UjNN5CtJ1tH1krWQraWum6m5mvIVged9ehlFXW2kmquIFtOtoxsKdkSatdE tphGtoiaN5I1UM16sjqyWrIasoVkC2jS82lk88jm0qSrqesqeqFKsrNpuHPohQLUSwVZOVkZWWkw 2o+JzQ5Gq8s6KxitXrAzg9E7YDOC0Vmw6VSlhGxaMBo/SEjFFE0lm0LJomD0VpQVBqN3wgqC0dtg +cHodtjkYGQRbBKZnyyPbGIwEj8XSGdRNCHoqEI0nmxc0KFeR2PJcoOOKYjGBB2VsNFBRzVsFJWN JMsJOjKRHEE1hwcd6sSGBR3qDSmbbCg1z6JXyCTLoM6GkKVTZ4PJ0sh8ZKlBh7pKg8hk6jOF+kym zrzUi4csidolkiWQucniyeKC9vnoMzZoXwCLCdoXwlxkTrJosiiySGrgoAZ2SkaQhZPZyKxU00I1 wygZSmYmM5GFUE0j1TRQUk+mI5PIhL8/os6jqi+i3tMb0eD5HfwbdAL6D3K/IvcL9DP0E/Qj8j9A 36PsO8THoW+hb6BjyH8NfYWyfyP+EvoC+hf0efhiz2fhTZ5PoX9C/4A+Qe4o/GPoI+hDxB/A34fe g/4OvWtb5nnHNtzzNvxvtuWet2w+z5vQG+DXbRme16Aj0KsofwW5l20rPC+BXwS/AH7ettTznG2J 51lbk+cZ22LPYbR9Gv09BT0J+fsP4fkJ6HHoMesaz6PWZs8j1rWeh60tnoNQD/QQ8g9CB1C2H2Xd yAWhLkiB9lk2eh6wbPLcb9niuc/S5um0bPXcC90D3Q3dBd0J3WHJ8twOvw26FW1uge+1LPPcDL4J fCN0A/h69HUd+roWfV2D3NXQVdCV0BXQ5dBlaHcp+tsTNtOzO2yW55KwxZ5dYXd4Lg67y3OePtVz rj7Xs0PK9WwPtAfO6WwPbAu0BbZ2tgUsbZKlzd1W0ra5rbPt/TZ/ZEjYlsCmwObOTYGNgfWBDZ3r Aw/rzheLdOf5JwTWdbYGDK3RrS2t+h9bpc5WqaBVGtYq6USrvdXbqre2BJoDazubA6J5dnN7s9Js GK80H23WiWYprKf/UHezO6kI7t/SbLMXrQmsCqzuXBVYuWhFYCkGuCR3caCpc3FgUW5DoLGzIVCf Wxeoza0JLMydH1jQOT8wL7c6MLezOlCVWxk4G/Xn5FYEAp0VgfLc0kBZZ2lgVu7MwEzkZ+SWBKZ3 lgSm5U4NFHdODUzJLQoUYvIiwZ7gTdDb1QHMTMBIhFuaPMztdx91H3cbhFtxH3LrIyPiPfG69Ig4 KX9WnLQqblvc7jh9ROyRWJ0/Nj2zKCLmSMzHMd/GGKL8MelDi4TL7vK69E51bq4ZFercul15BeTD R2lzneGSfUURTinC6XHqCj1OSTiOOo479M4n7EfsuogIKSKiP0Lnj0D1iHBPuE596g/X+8OHjymK sHlsOvWp36Z3+W3IqINPs86uKIqweCy6QJ5llkXnt+TlF/ktWcOKhF7ySvifHztMb1ZHIzk9RT2S 6HZJRqlH2tNVUZ6RUdJjFmUlinn2XEW6QEktV5/9pdVKyAWKCFTPreySpEuquiRdfoUSXVJaTfF5 u3aJyYklSmJ5pbI3sapEaQf4VegHiMQul5hclbFgbevajIyWBXhasLYlQ/tGJLWqER4owPfaFsTq FwyxUEv+/EHVUG/hWjy0bqj3P2/yf6BE+j8wxr/4ELsEjmjlpH7duaJBtwPaDp0DtUPboK1QG7QF 2gxtgjZCG6D10DqoFWqB1kJroNXQKmgltAJaDi2DlkJLoCZoMbQIaoQaoHqoDqqFaqCF0AJoPjQP mgtVQ1VQJXQ2NAcKQBVQOVQGlUKzoVnQTGgGNB0qgaZBxdBUaApUBBVCBVA+NBmaBPmhPGgidBY0 ARoPjYPGQrnQGGg0NAoaCeVAI6Dh0DAoGxoKZUGZUAY0BEqHBkNpkA9KhQZBMpQCJUNeyAMlQYlQ AuSG4qE4KBaKgVyQE4qGoqBIyAHZoQgoHLJBVsgChUGhkBkyQSGQETJM6sezHtJBEiREg4Sc1Af1 Qr9Dv0EnoP9Av0K/QD9DP0E/Qj9A30PfQcehb6FvoGPQ19BX0L+hL6EvoH9Bn0OfQZ9C/4T+AX0C HYU+hj6CPoQ+gN6H3oP+Dr0LvQO9Df0Negt6E3oDeh16DToCvQq9Ar0MvQS9CL0APQ89Bz0LPQMd hp6GnoKehA5BT0CPQ49Bj0KPQA9DB6Ee6CHoQegAtB/qhoJQF6RA+6AHoPuh+6BO6F7oHuhu6C7o TugO6HboNuhW6BZoL3QzdBN0I3QDdD10HXQtdA10NXQVdCV0BXQ5dBl0KbQH2g1dAu2CLoYugjqg C6ELoJ3Q+dB5omFSu3QuaAe0HToHaoe2QVuhNmgLtBnaBG2ENkDroXVQK9QCrYWaoTXQamgVtBJa AS2HlkFLoSVQE7QYWgQ1Qg1QPVQH1UI10EJoATQfmgfNhaqhKqgSOhuaAwWgCqgcKoNmQ7OgmdB0 qASaBhVDU6EpUBFUCBVA+aLhL36b/qsPr+qvPsC/+PhiFy5Qf2NIiL7LT/8lITFbLBVrRTu+zhe7 xOXiCfG+qBM7QNeKveJOcY9QxJPiBfHOGa3+l0HfRuMKYdU/JEJElBD9J/qP9d0J9RjDT8tcjijK 4D2V6bf3f/OH3Dd9l/fb+3pCIkWY1tamewO9/SD19p/A+2uIsPWPVmPdTnCE9krfmW7q29d31xkT mC1KRbWYK+aJ+aJG1GL+DaJJLMHKLBPLxQqxUotWomwxeBGihaiFe4nGp2qtEqvFKtEsWkSrWIev 1eC1A5FatkaLW8V6fG0QG8UmsVlsEW0Dz+u1zBaUbNKyG1CyVWzDzpwjtmvETpkd4lxxHnZtp7hA XIgd+/PowpO1OsRF4mLs8yVit/gz3nVGyR6xR1wqLsN5uEJcKa4S1+BcXC9u+EP2ai1/nbhJ3Iwz o7a4EpmbNbpKXC0eFc+KA+IBsU88qK1lPdaWVoTXZZG20quxBlsw5x2njZhWc/3J1dqK1VDn3TEw 7w1Yv+2ntVg3sI7q6u1ATXV1Ogb2Qe2lbSDDK7EHMyM+NU91jdQ57D5jntzif8qqM1bX6QasF6+M umZXIXfd/5c9vcbpfJW4EVfgLXhWV1WlW8FEN2t8ev6mk3X3amW3idvFHdiLu4RK7JS5E7m7xN24 tu8VneI+fJ3i04lKHxD3azuniC4RFN1iP3byQfGQ6NHy/13ZPtw7/time6Cv4MleDoqHxSM4IY+L Q7jTPIUvzjyG3BMD2cNaLYqfwu9SHtZqqaVP4Ww9hzvUi+Il8bI4Ip5B9Kr2/Dyi18Qb4k3xjmQD vS6+xHOveM34KX41cxJ+8fJh7MYNYoFY4J/SsHDB/Hlzq6sqAxXlZaWzZ82cMb1kWvHUKUWFBfmT J/nzJp41Yfy4sbljRo/KHpqVOdiXOkhO8cRGO+wRNktYqNkUYjTo8ZNtZqFcVONVfDWKwSdPnZql xnItErWnJWoUL1JFZ9ZRvGq7WhSdUdOPmov+UNNPNf0na0p27wQxISvTWyh7lVcKZG+PVF1aCd5V IFd5lWMaz9DY4NMCG4LkZLTwFsY2FXgVqcZbqBSta+oorCnIypS6LGH5cn5jWFam6AqzAC0gZbC8 uksaPFHSQDe4cFyXTpht6ssq+tTC2gZldmllYYE7OblKy4l8rS8lJF8xaX15lygYs7jI25V5qOPi Hruoq8mwNsgNtfMqFX0tGnXoCzs6diqODCVdLlDSN30aiwVsVDLlgkIlQ8bASspOvoCkGFPtsrfj J4HBy8e+xqhPy9QOZEJS7T8JtVCd4sllUqRaZoGxYYSYX3KyOpaLevyiDoHSXlpJsVfUuYPCn51R pehq1JJDXOIMqCXtXHKyeY2MlS2UC2sGvtc1xSrtdd6sTOys9p2qGFJR7lX0vpq6+ibVaxs75ALM EGspKvChTQHAXzuwmIVdw7JRv7YGk1iiLkNppZItr1ai5cm02kigk9TCJeWVWhPKFirR+YqoqR9o pWQXoi2OSGGHujHqANW+5NLKgyKn/2jXSK+7O0eMFFXqOBRXPjbFV9hR2bBI8dS4G3A+F3kr3cmK vwrLVyVXNlapuyTblfSjeDk8sIFaK8ztD7W5MqatmFLN3kqdW1+l7hYS3iI8yZMnoMCuhFCo7ujk Cd5KyS24Gl5loIZKZ/SDQJ+aPxWN4WiaP9WdjMOtPf6bIblpAhiGYj45JgMGYTw1JnqdPx0a1VYH lO4tbCw4bYBndIpAG+BAb//1OHXqWgwsBoZgVrdzqjqHrEwd2Itis6LDPLWUuouxXkXM9lbKjXKV jDPkn12pbo661tr+lpTL6geD2m4PnJKKMyIqz6UyRSSXVFRyoH5moxRlaPuqbqsWT9Hik+HUPxQX c7G3wyyXlHeoLy4PdCi8uIKwOSG+4tqLciNH4mItwo1SLqqVvXZvUUdtT397XUeX39+xurCmaRwu gw65uKFDLq+cgL3Urvs29yb1pSNFiVRSMTkrE/eeyV2ydEFpl1+6oLy68iD+AMB7QUVlUIcPRWsm V3UNQlnlQa8Qfi2rU7NqUq3iVQO1pzIEZq2++6BfiHat1KAltLgen8tqOaqEnCTqe3SUs3M9HXIG yvm1XBUeuMJim7AFuA8XehvU7dlS1dRRU6VeXMKFrcS3pEjyRKHo5In4KDfEqoTJjZMVizxZzeep +TzKh6h5kzxZkVwSFqcH96SOGhn3KRy5SnxEXoXTYVdPvy7V29PfX1GZ/Ir7WFUyLol5UHWlEpqB 9wFj6jTUm6KqBukpSnt9rToOEcClrl6ZxfVVuBa4Q1QpVkLRQ+hAD6hRpLVRjyMa1WNvsIFa+3YE SnuVUpWhvmjlEnVEXq9dEVPlcdh26tPoU18ou6ojUh6hHmxUVcJSd6oWirEJfEitZdwI8WK44aoz Mlkx8noZRfU1XuyAQdSX46jTvTRM3TdkGnFLNPgaNYW5BwqFOi19qsUWpoQORYf4VtkyFB3i21SF RVEnr0U7Byrgte2KBSPynbaUAw2wOigqVseC750YvFr1SbWb0h5RJm/ArVEdtPZSJhQrttTiWtz8 qb0FGTmXG6Mvc6qaUvs4TFmTOnMr1l2fWtHTf5e8Ub0D8CMrU1bfHNSDKdwHcbBFVccfE8rcjKxM 8x+zNi3d0WG2/dcNaL3MtpOu9uItxHuNEAb1z1iOwG8RsqFa3GcoELWGr8V9+i+g+xGniWH6z0WS /gkxHgOjv2kRwop/cyUhThZhQoesCbtkxN/4hCAOFfRnLreIW6Rs6TNdne6IvkS/35BguNjwLmqJ vrX6N/AvOj1ajRUzxEwx91Fhw0cvLjFOOnDAWVBgzjI9jo9VdMKLD2bMQpLy/REGne2h+Pg8+aFR Ibv0juIeKWt/nmkXPnLM6/2o99Xs3o+ORY7NPiZlf/jJR5/Yv3vVMTY755O3Phk+THIkOzRFh+tM pugQOWWoblSab3ROzoiJulEjfXJKuE7LjRw9ZqI+Z0SSTo+alJmoU2NJ/8bv1fpZvSG6rXLenBxj UnxEtC3EqEuIjcyakGovn5s6YWiiSW8K0RvNpsFjJqeULC9Mec/kSHS6EiPN5shElzPRYep93xh+ 4ntj+G/5huW/XaEPGT8vb5D+mjCzzhAS0pMUGzdkfHLxnIgou8ESZXe4zKZIh3Vwwbze850Jah8J Tif11TsDyyL3nzBsNUaLFOETNx4Ug/q/2G+1S9PlngHw9fQf329BxsKA//s77o9XU6l29dmmPVu1 Z/9gKVUtzrRIMwbJvtQfrRZrbEqiHGaTXAarsNqtun3yE/IRWS9bZWtkYllkwBgQeXl5kWPHZmfP n++IGesAOnLsx0Y4crDiGfPp8xL8r1KqyxWiLXmaPlkfrpdTfL7RYyRa5xiTrE82tJole6rHkxoV aljV+/lSfViUnJCYGiGZpaDBFpeW5B0SH27YLH0sPXWWyx1u0JusodL4vhdCbaEGY7jbZQhaws16 vTnCsqt3s8CZug9nWcLpShIZIlc874/3xNqlGR57hPpkw1OsFU9ezBW/iTbUPzje6Ue5049yp9OS qVbOVCtnqpUz1cqZauXMh3Uj8BnDoQNg4cvBSnejJvx4Nyprjvrwn7vRRCtHzZwend1v22s5ZNFZ 4tN+HD7cNKhHwm8IlI7skSxdpgqRdyxPO7djpez5n2irNuKtDAIc54yMscRY1Ohwg5yc4hvlGDk6 Jxmn0qme5yS9NHKoTpYd6mGOOoUGyZM7q35Ncd8DMenpMZKv5Yr6Ea6MSUNGzSsc3Ncbn1s9LXg4 v2x03MzUKctKXz0xvjLfJ609a3HZxCFOT5phe5ons2LTjKEVU3Ijw0aVrdRJ2dNHJfTNl8fP6v1w XOUET19uwpgyXPW1/ccNVmMSruK67gQxPmNgVeDaqsC/7saqwL9RV0Urx6pkPK7Lwb/RYqVs3D98 UmYwqtzwiDREjBLDpKFdoXNwSb91TJWUTdO3v314+LDU6HC6eEdql2WIugDqZapewM7oJFyqdLka rDqjOdq/cHPx1pd2zyi/6vVtuUuri9xmo95gtpjDR8xaM2vOroYxo+r3zJ2xtnRkhCksRP+QPTYy PDo9zV1x+3c33vL7vnlO7xB3eFR8ZHRCVGhadlrh+U9u2fzYtkm+bF+IIwlXoHrKduOURQqPWO9P zEuWotSTE6WenKhozDkqEhOOisVsox5RT46IpxMTP7A2mqMeXDsx8C+6UTv+EfwfTCjWxhoML3X3 SL4uI50SXou3+ETMx5HQnXEkTKcdgN1z7jh+Z9832van3v3FjaUHRq669/x9XVvubR6ru+7u3+4o o40++7Yvrl1y4Nxpvzsmtj+p3t8xM/0WzCxTrOuKTxvYUbi2o5pj1HBt1Fo5Rp3Wo3P4Q0OjvFFe DD6+RzL7be0+6ZBPes0n+XwhcT2Yj600DdYVcvLUz1/TjC3O1m4jdprWCG2fz5yWttHJDj70MqF+ iyHMZu69XJ2hbpHZZjYa8dQXIgXNuDUYQsEzdZLZFmaYEumONNNszZHu6Ei3w9y3NNSeEBUZbzf1 DTc73Nq8+0/oKzDvNDGvyxQ1MG+4Nm/N1V0dmLdWru4t5n3AliiSEk2YWndUVFxIjzS4O6U0Tr1B DrwjZR92DFzGmB1t2umT4XcbvrD1FZiYqQ+rZ8LgNfabo73xsSnRZky1SMsejkrALKaa7G5nlNsR 2vuZyWYyGvFkeCDNg7ccdUbDMKOf8S6RLYb5E9OzpfShki9W8sVIaS5psJDSy2SLI7HMcepmnofh zdcep27bknTyro2NHD1afUOMkbQ7OG7tJkn/qc0YmZ7iHeS0GPqO9n1otDoHJSX7Iow2qbZvn9Vk Hywn+VxhIfjRNtoYFpWS6ElzGKx9ykRXfIRRb7aE6vS9vaFWk94YEe/SlevyXO4I3OSxewnSp2Yb 8rjJ9z6jzicJ83kO8xkiqvDjEV1OYmCDNMcGwbWDCdcuJ/EwLicrfgsjRpphVX9zw4pK1tQyd0hk WYi6R5Fj8yTcZ/mikhwnt+PUdKNcrpic0aPHRKnvW+qsdcXm6JTYuGSnue8yizEiLTkp1WUxdseN iNfFDI/br7dEpcQPSrcbLdIvfbLZqu6M1Sx9qHsvBu9lBmxg38WjWsaPXTNGWhcWbsL7WLwL88PP V/orcALTRXxXigOXfre71IpLJo1uASPyjp1+dYzCbXCMlKz+NKPeEJNHuJzaW4IE0l+B8x/aa4j3 6sMibbpAbzAsXL0mwsN0r7k9hjBHeO8Dug2OyKlRuDK8cqrNFedx6u/ElYAbHq4Vb5o9Lj4p+vcF KbhxYWSRkPoIUf/ouSQwJzCzICO/dvmSuuYl/w/p8oW1CmVuZHN0cmVhbQplbmRvYmoKNTYgMCBv YmoKNzYxNgplbmRvYmoKNTcgMCBvYmoKKHRvcG8xLnBwdHgpCmVuZG9iago1OCAwIG9iagooTWFj IE9TIFggMTAuOS41IFF1YXJ0eiBQREZDb250ZXh0KQplbmRvYmoKNTkgMCBvYmoKKEFsYmVydCBU aWFuKQplbmRvYmoKNjAgMCBvYmoKKCkKZW5kb2JqCjYxIDAgb2JqCihQb3dlclBvaW50KQplbmRv YmoKNjIgMCBvYmoKKEQ6MjAxNDExMTgxOTEzMjBaMDAnMDAnKQplbmRvYmoKNjMgMCBvYmoKKCkK ZW5kb2JqCjY0IDAgb2JqClsgKCkgXQplbmRvYmoKMSAwIG9iago8PCAvVGl0bGUgNTcgMCBSIC9B dXRob3IgNTkgMCBSIC9TdWJqZWN0IDYwIDAgUiAvUHJvZHVjZXIgNTggMCBSIC9DcmVhdG9yCjYx IDAgUiAvQ3JlYXRpb25EYXRlIDYyIDAgUiAvTW9kRGF0ZSA2MiAwIFIgL0tleXdvcmRzIDYzIDAg UiAvQUFQTDpLZXl3b3Jkcwo2NCAwIFIgPj4KZW5kb2JqCnhyZWYKMCA2NQowMDAwMDAwMDAwIDY1 NTM1IGYgCjAwMDAwMzgwNDkgMDAwMDAgbiAKMDAwMDAwMTI5MSAwMDAwMCBuIAowMDAwMDI5MTAy IDAwMDAwIG4gCjAwMDAwMDAwMjIgMDAwMDAgbiAKMDAwMDAwMTI3MSAwMDAwMCBuIAowMDAwMDAx Mzk1IDAwMDAwIG4gCjAwMDAwMjkwNjYgMDAwMDAgbiAKMDAwMDAyNzg1OSAwMDAwMCBuIAowMDAw MDAwMDAwIDAwMDAwIG4gCjAwMDAwMjkyNDkgMDAwMDAgbiAKMDAwMDAwNjgxNCAwMDAwMCBuIAow MDAwMDA3Nzc2IDAwMDAwIG4gCjAwMDAwMDMzNTggMDAwMDAgbiAKMDAwMDAwNDg5MiAwMDAwMCBu IAowMDAwMDAyMDAyIDAwMDAwIG4gCjAwMDAwMDIzNDMgMDAwMDAgbiAKMDAwMDAwMjM2MyAwMDAw MCBuIAowMDAwMDAzMzM4IDAwMDAwIG4gCjAwMDAwMDQ5MTMgMDAwMDAgbiAKMDAwMDAwNTI1MSAw MDAwMCBuIAowMDAwMDAxNjUxIDAwMDAwIG4gCjAwMDAwMDE5ODIgMDAwMDAgbiAKMDAwMDAwNzc5 NiAwMDAwMCBuIAowMDAwMDA4MTI3IDAwMDAwIG4gCjAwMDAwMDgxNDcgMDAwMDAgbiAKMDAwMDAw ODQ3OCAwMDAwMCBuIAowMDAwMDA1MjcxIDAwMDAwIG4gCjAwMDAwMDY3OTMgMDAwMDAgbiAKMDAw MDAxMzMxOCAwMDAwMCBuIAowMDAwMDE0MDI1IDAwMDAwIG4gCjAwMDAwMTI1NzAgMDAwMDAgbiAK MDAwMDAxMzI5OCAwMDAwMCBuIAowMDAwMDIyNTE4IDAwMDAwIG4gCjAwMDAwMjUxMDIgMDAwMDAg biAKMDAwMDAxNzI4MCAwMDAwMCBuIAowMDAwMDIxMTIwIDAwMDAwIG4gCjAwMDAwMTQwNDUgMDAw MDAgbiAKMDAwMDAxNDcwMiAwMDAwMCBuIAowMDAwMDA4NDk4IDAwMDAwIG4gCjAwMDAwMTI1NDkg MDAwMDAgbiAKMDAwMDAxNDcyMiAwMDAwMCBuIAowMDAwMDE3MjU5IDAwMDAwIG4gCjAwMDAwMjE3 OTcgMDAwMDAgbiAKMDAwMDAyMjQ5OCAwMDAwMCBuIAowMDAwMDIxMTQxIDAwMDAwIG4gCjAwMDAw MjE3NzcgMDAwMDAgbiAKMDAwMDAyNTEyMyAwMDAwMCBuIAowMDAwMDI3ODM4IDAwMDAwIG4gCjAw MDAwMjc4OTUgMDAwMDAgbiAKMDAwMDAyOTA0NSAwMDAwMCBuIAowMDAwMDI5MTg1IDAwMDAwIG4g CjAwMDAwMjk4NDMgMDAwMDAgbiAKMDAwMDAyOTQ0OCAwMDAwMCBuIAowMDAwMDI5ODIzIDAwMDAw IG4gCjAwMDAwMzAwNzggMDAwMDAgbiAKMDAwMDAzNzc4NSAwMDAwMCBuIAowMDAwMDM3ODA2IDAw MDAwIG4gCjAwMDAwMzc4MzUgMDAwMDAgbiAKMDAwMDAzNzg4NyAwMDAwMCBuIAowMDAwMDM3OTE3 IDAwMDAwIG4gCjAwMDAwMzc5MzYgMDAwMDAgbiAKMDAwMDAzNzk2NSAwMDAwMCBuIAowMDAwMDM4 MDA3IDAwMDAwIG4gCjAwMDAwMzgwMjYgMDAwMDAgbiAKdHJhaWxlcgo8PCAvU2l6ZSA2NSAvUm9v dCA1MSAwIFIgL0luZm8gMSAwIFIgL0lEIFsgPGZkYzFkMWFlMGJmY2VlMWZlMjViMGZjNTFiOTQ5 YzZlPgo8ZmRjMWQxYWUwYmZjZWUxZmUyNWIwZmM1MWI5NDljNmU+IF0gPj4Kc3RhcnR4cmVmCjM4 MjI0CiUlRU9GCg== --_002_D090DCA031617AAlbertTianericssoncom_-- From nobody Tue Nov 18 12:05:28 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 525C81A8749 for ; Tue, 18 Nov 2014 12:05:23 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.902 X-Spam-Level: X-Spam-Status: No, score=-1.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id d7HXxDefoXYW for ; Tue, 18 Nov 2014 12:05:13 -0800 (PST) Received: from na01-bn1-obe.outbound.protection.outlook.com (mail-bn1bon0712.outbound.protection.outlook.com [IPv6:2a01:111:f400:fc10::1:712]) (using TLSv1 with cipher ECDHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3A81A1A874B for ; Tue, 18 Nov 2014 12:05:12 -0800 (PST) Received: from CY1PR0501MB1100.namprd05.prod.outlook.com (25.160.144.142) by CY1PR0501MB1164.namprd05.prod.outlook.com (25.160.145.13) with Microsoft SMTP Server (TLS) id 15.1.16.15; Tue, 18 Nov 2014 20:04:49 +0000 Received: from [172.29.32.63] (66.129.241.11) by CY1PR0501MB1100.namprd05.prod.outlook.com (25.160.144.142) with Microsoft SMTP Server (TLS) id 15.1.16.15; Tue, 18 Nov 2014 20:04:48 +0000 Message-ID: <546BA65A.9090002@juniper.net> Date: Tue, 18 Nov 2014 15:04:42 -0500 From: Eric C Rosen User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.2.0 MIME-Version: 1.0 To: AshwoodsmithPeter , "bier@ietf.org" References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D208@dfweml703-chm> <546B6938.9020709@juniper.net> <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D531@dfweml703-chm> In-Reply-To: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D531@dfweml703-chm> Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit X-Originating-IP: [66.129.241.11] X-ClientProxiedBy: BLUPR11CA0082.namprd11.prod.outlook.com (10.141.30.50) To CY1PR0501MB1100.namprd05.prod.outlook.com (25.160.144.142) X-Microsoft-Antispam: UriScan:;UriScan:; X-Microsoft-Antispam: BCL:0;PCL:0;RULEID:;SRVR:CY1PR0501MB1100; X-Exchange-Antispam-Report-Test: UriScan:; X-Exchange-Antispam-Report-CFA-Test: BCL:0; PCL:0; RULEID:; SRVR:CY1PR0501MB1100; X-Forefront-PRVS: 039975700A X-Forefront-Antispam-Report: SFV:NSPM; SFS:(10019020)(6009001)(6049001)(189002)(199003)(95666004)(105586002)(102836001)(33656002)(50466002)(106356001)(4396001)(64126003)(40100003)(101416001)(21056001)(46102003)(107046002)(97736003)(87266999)(2501002)(65816999)(76176999)(36756003)(120916001)(99396003)(54356999)(77096003)(92566001)(59896002)(122386002)(92726001)(20776003)(64706001)(77156002)(31966008)(65806001)(42186005)(23746002)(50986999)(87976001)(83506001)(66066001)(62966003)(80316001)(86362001)(47776003)(65956001); DIR:OUT; SFP:1102; SCL:1; SRVR:CY1PR0501MB1100; H:[172.29.32.63]; FPR:; MLV:sfv; PTR:InfoNoRecords; MX:1; A:1; LANG:en; X-Exchange-Antispam-Report-CFA-Test: BCL:0; PCL:0; RULEID:; SRVR:CY1PR0501MB1100; X-Microsoft-Antispam: BCL:0;PCL:0;RULEID:;SRVR:CY1PR0501MB1164; X-OriginatorOrg: juniper.net Archived-At: http://mailarchive.ietf.org/arch/msg/bier/GQhKYUXwhtX2soNgZZLAPyo0Gs0 Cc: erosen@juniper.net Subject: Re: [Bier] Follow up on my comments from Honolulu concerning T.E/ Steiner Tree optimization with Bier X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 20:05:23 -0000 Peter, I don't think we have a substantive disagreement, but I do like to distinguish between "X is prohibited by the BIER architecture" and "combining X with BIER is a bad idea". Perhaps I'm being overly pedantic ... > the computation of the next hops has no knowledge of all the intended > member subsets. The BIER architecture does not say what facts may or may not be known when the next hops are computed. (That's the pedantic point ;-)) > the MT computed tree creates state for the use of only that set of > destinations and is essentially removing the stateless (per set of > destinations) nature of Bier. This is the substantive point. The more pre-computed trees you have, and the more specific each one is to a particular set of egress nodes, the less benefit you get from BIER. I think that's absolutely correct. If you wanted a Steiner tree for each flow, you would not use BIER. > just trying to be clear on what it can and cannot do and how it would > have to be used to get results similar to say P2MP central control. Certainly global optimization of the multicast flow patterns is unlikely to be achieved with a solution that is both (a) distributed and (b) low in multicast-specific state. Eric From nobody Tue Nov 18 12:15:11 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 364C41A8747 for ; Tue, 18 Nov 2014 12:15:07 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.201 X-Spam-Level: X-Spam-Status: No, score=-4.201 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0_1u1zVzWtzs for ; Tue, 18 Nov 2014 12:15:04 -0800 (PST) Received: from usevmg20.ericsson.net (usevmg20.ericsson.net [198.24.6.45]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9706F1A86FE for ; Tue, 18 Nov 2014 12:15:04 -0800 (PST) X-AuditID: c618062d-f79206d0000014d2-a1-546b4fbb7932 Received: from EUSAAHC002.ericsson.se (Unknown_Domain [147.117.188.78]) by usevmg20.ericsson.net (Symantec Mail Security) with SMTP id ED.CB.05330.BBF4B645; Tue, 18 Nov 2014 14:55:08 +0100 (CET) Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC002.ericsson.se ([147.117.188.78]) with mapi id 14.03.0174.001; Tue, 18 Nov 2014 15:15:02 -0500 From: Albert Tian To: AshwoodsmithPeter , "bier@ietf.org" Thread-Topic: [Bier] BIER and deterministic ECMP Thread-Index: AQHQAvPmimRcG0ie5kmRBDyo9YL80ZxmvLFw///jxoA= Date: Tue, 18 Nov 2014 20:15:02 +0000 Message-ID: References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4EE@dfweml703-chm> In-Reply-To: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4EE@dfweml703-chm> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.4.140807 x-originating-ip: [147.117.188.9] Content-Type: text/plain; charset="us-ascii" Content-ID: <60F0F3BE55C5BF48B8E3CC504C0E733C@ericsson.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFvrBLMWRmVeSWpSXmKPExsUyuXSPn+4e/+wQg+07xS2WztjDZLHwbhuT A5NHy5G3rB5LlvxkCmCK4rJJSc3JLEst0rdL4MrYOu0uW8EWpYoDp24yNjBukOli5OSQEDCR eHh/CxuELSZx4d56IJuLQ0jgCKPErlttzBDOckaJD3OfsYBUsQnoSLya/YARxBYRiJDoudMM ZgsL6EvsO/qTvYuRAyhuINHdxAVRYiXRsOgaM4jNIqAqMf1DB9gYXgEziWO3DoK1CgkUSvS9 /cIOYnMKuEq8+noE7CBGoIO+n1rDBGIzC4hL3HoynwniUAGJJXvOM0PYohIvH/9jBbFFBfQk nm3YzA4RV5TY1z+dHaJXR2LB7k9sELa1xL9LPVC2tsSyha+ZIe4RlDg58wnLBEbxWUjWzULS PgtJ+ywk7bOQtC9gZF3FyFFanFqWm25ksIkRGFXHJNh0dzDueWl5iFGAg1GJh3eDa1aIEGti WXFl7iFGaQ4WJXHeWbXzgoUE0hNLUrNTUwtSi+KLSnNSiw8xMnFwSjUwVurYuMrL+3peNzD+ /WV12ZM513znP9r7g3fNPZn3535V68Y/eNaUtL78yAfZ9bm5jTsjlZltu/jLeSx+NiqZ5Z6W 8vjIn3ZYU+XCsQ0T1s1c6il8vemCWcvarSk54duO3u+X+XvrpfMS9U1drFaq367PDFebd3D+ Ge2vVxNlmp4JP++bkJ92VImlOCPRUIu5qDgRAIWO4IeLAgAA Archived-At: http://mailarchive.ietf.org/arch/msg/bier/0dzYTy1ialX7ltQGYqhyZxRZ8dw Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Nov 2014 20:15:07 -0000 Hi Peter, If I understood it correctly, your proposal is still non-deterministic. Consider fluctuation of group members behind the lower ordered links, they could still impact the paths of the members behind the higher ordered links, assuming you are going to have F-BM on for each link. Statistically speaking, lower ordered links would also draw more traffic due to the F-BM operations. Thanks, Albert On 11/18/14, 11:00 AM, "AshwoodsmithPeter" wrote: >What if rather than process in the order of the bits the processing is >inverted and instead the links are processed in some lexicographic >ordering as the outer loop. That would lead to a determinism of the paths >used irrespective of entry or departure of lower order bits. Slower I >guess for a sparsely populated bit set. > >Peter=20 > >-----Original Message----- >From: BIER [mailto:bier-bounces@ietf.org] On Behalf Of Albert Tian >Sent: Tuesday, November 18, 2014 12:53 AM >To: Ross Callon; bier@ietf.org >Subject: Re: [Bier] BIER and deterministic ECMP > >Hi Ross, > >Very good point. Yes IMHO this needs to be addressed, otherwise the links >leading to lower numbered BFERs will more likely be chosen in ECMP and >cause unbalanced traffic. This maybe just as bad as the instability you >have identified. > >Agreed multiple BIFT solution should work. > >Thanks, >Albert > > > > > >On 11/17/14, 3:10 PM, "Ross Callon" wrote: > >>Reading the BIER architecture (draft-wijnands-bier-architecture-01) I >>noticed the last sentence of section 6.7: >> >> Note however that by the rules of Section 6.5, any packet destined >> for both BFER-D and BFER-F will be sent via BFR-C. >> >>Thinking about this more: It is clear that where ECMP paths exist, >>which path is taken to reach any particular destination will depend >>upon which other destinations the same packet goes to. In particular, >>which lower numbered bits are set in the BitString can change the >>choice among ECMP paths. This could have the effect that as a lower >>numbered destination intermittently joins and leaves a multicast group, >>the path to a higher numbered destination could switch back and forth >>between different ECMP paths. >>=20 >>This could be fixed if we used multiple BIFT tables, such that each >>BIFT table has only one path to each destination, but the multiple >>possible ECMP paths to any particular destination are spread across the >>multiple tables. When a BIER-encapsulated packet arrives to be >>forwarded, the BFR would use the BIER Entropy field (or a hash thereon, >>if fewer than 256 next hops are >>supported) >>to determine which BIFT to use, and then the normal BIER forwarding >>algorithm (the same as described in >>draft-wijnands-bier-architecture-01) would be used with whichever BIFT >>table was selected. This would of course increase the amount of memory >>that would be needed to store BIFT tables, but my quick "back of the >>envelope" calculation suggests that the amount of memory to store BIFT >>tables still should not be a problem if there aren't too many BIFT >>tables (I did my "back of the envelope" with max 16 tables). The >>forwarding algorithm and packet headers stay unchanged. This also has >>the feature that if this approach is consistently implemented across a >>service provider, then where a problem has been reported on a >>particular flow of packets to a particular receiver, we can traceroute >>with the same entropy value to that particular receiver and know that >>we will be using the same path as the normal data packets to that >>receiver. >>=20 >>I would be happy to write up appropriate text to add to >>draft-wijnands-bier- >>architecture-01 (presumably for a -02 version) if the authors would >>like, or alternately could write up a separate Internet Draft to >>further describe the issue and the solution. >> >>Thanks, Ross >> >> >> >>_______________________________________________ >>BIER mailing list >>BIER@ietf.org >>https://www.ietf.org/mailman/listinfo/bier > >_______________________________________________ >BIER mailing list >BIER@ietf.org >https://www.ietf.org/mailman/listinfo/bier > >_______________________________________________ >BIER mailing list >BIER@ietf.org >https://www.ietf.org/mailman/listinfo/bier From nobody Mon Nov 24 08:23:16 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B39CC1A86F6 for ; Mon, 24 Nov 2014 08:23:05 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.999 X-Spam-Level: X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9dfX62tBIyWM for ; Mon, 24 Nov 2014 08:23:01 -0800 (PST) Received: from mail-wi0-x22c.google.com (mail-wi0-x22c.google.com [IPv6:2a00:1450:400c:c05::22c]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 3F1231A8549 for ; Mon, 24 Nov 2014 08:22:53 -0800 (PST) Received: by mail-wi0-f172.google.com with SMTP id n3so6247516wiv.5 for ; Mon, 24 Nov 2014 08:22:52 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:reply-to:date:message-id:subject:from:to:content-type; bh=0Fqpg64SFbmpkzewILs2LuTXFMgUwN8OfkmkNGnwqFQ=; b=Zn9/foVOEKZJO9354gya5N/kaQnRK3tq371qIQZD7lZLzePEq/pZM6Sn28BJEpF653 q6QGGoAHuK/gAJEj2w4ITug0ezCcS6i3CnSDmuvdMH4nCBmjdTyL2aZ8BP758Fzw0HZ8 VuC5zdOUuUX/zBneLkno9eYr/QfhvoZIbbI1yVVAs0RCPcbGsczV14nFSr1ywmjFrE3b BZ+kE9qkzTOk8EJnHQzSIHjBiWcxb0/gHaFZo6V/r4n4E+1Tjd9TC++2vqGTSi+DDKIH 0oV8BuH8BLi+orVCHbhb/WvuLf0ZxBl4frGTDwa65PmmMNJB96CEQw7Wu8UamrBsFEJO jlMw== MIME-Version: 1.0 X-Received: by 10.180.73.235 with SMTP id o11mr4644372wiv.51.1416846162389; Mon, 24 Nov 2014 08:22:42 -0800 (PST) Received: by 10.180.207.200 with HTTP; Mon, 24 Nov 2014 08:22:42 -0800 (PST) Date: Mon, 24 Nov 2014 08:22:42 -0800 Message-ID: From: Greg Shepherd To: "bier@ietf.org" Content-Type: multipart/alternative; boundary=f46d043d66dd35ee5c05089d346e Archived-At: http://mailarchive.ietf.org/arch/msg/bier/vcNJURO_5M3LV2fpHbM3j8Dds98 Subject: [Bier] Virtual Interim Meeting X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gjshep@gmail.com List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Nov 2014 16:23:05 -0000 --f46d043d66dd35ee5c05089d346e Content-Type: text/plain; charset=UTF-8 Thank all of you who participated in our first BIER BoF! I think the community response was overwhelmingly positive, but that just sets us up to get this work moving. I'd like to propose we have a virtual interim meeting to focus on the existing work, and assign editor/contributors for each draft - getting commitments to both get the work progressing AND to show the ADs that we have committed people to progress the work - and to review any new work that may not have been presented at the BoF. I know this is a difficult part of the year, but let's see if we can find a day/time that works for the majority. My proposed dates are: Dec 11, 12, or 15. I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I don't mind getting up at odd hours for a call if needed. Can each of the editors of the current drafts please speak up here as to which day/times work best for you. And anyone with new work that they wish to present please respond as well. Looking forward to working with all of you. Cheers, Greg --f46d043d66dd35ee5c05089d346e Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
Thank all of you who participated in our first BIER BoF! I= think the community response was overwhelmingly positive, but that just se= ts us up to get this work moving.

I'd like to propos= e we have a virtual interim meeting to focus on the existing work, and assi= gn editor/contributors for each draft - getting commitments to both get the= work progressing AND to show the ADs that we have committed people to prog= ress the work - and to review any new work that may not have been presented= at the BoF.

I know this is a difficult part of th= e year, but let's see if we can find a day/time that works for the majo= rity. My proposed dates are:

Dec 11, 12, or 15.

I'll be in EU on the 11th and 12th, and back in = PDT on the 15th. But I don't mind getting up at odd hours for a call if= needed.

Can each of the editors of the current dr= afts please speak up here as to which day/times work best for you. And anyo= ne with new work that they wish to present please respond as well.

Looking forward to working with all of you.

=
Cheers,
Greg
--f46d043d66dd35ee5c05089d346e-- From nobody Mon Nov 24 09:04:12 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0DD761A8737 for ; Mon, 24 Nov 2014 09:04:12 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.51 X-Spam-Level: X-Spam-Status: No, score=-14.51 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4nZ-cTchKDQS for ; Mon, 24 Nov 2014 09:04:10 -0800 (PST) Received: from alln-iport-4.cisco.com (alln-iport-4.cisco.com [173.37.142.91]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id C3C591A873E for ; Mon, 24 Nov 2014 09:04:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=4692; q=dns/txt; s=iport; t=1416848648; x=1418058248; h=from:to:subject:date:message-id:in-reply-to:mime-version; bh=ApyVix5PSICGyFuacONJj/byCZtX7D1FTmkdYZdQ85c=; b=El1vPXM7Qbywv2cTWEOcT2ish0qmKfmeYjBL9kVXazoMTpPdcv19xGhs ZjcZbVkWtINdJS8+pk6eSn1euETOFw9sC0zucMp0BzJA94TslFzEe8BLu o2vdwXYd1zRGKFjZNcnByqPf2NAxESXRQQs/Efxxi89kkFPzDiTkNz2sV k=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AjUFAFVkc1StJA2F/2dsb2JhbABbgkhGgS4EyGKJIwKBHhYBAQEBAX2EAgECBIELAQgEDQMBAigoERQJCAIEARIUiBgDEslODYZGAQEBAQYBAQEBAQEcjk6CLBiETgWQOoIuigSCFYE0g1mLD4Z+g314gUiBAwEBAQ X-IronPort-AV: E=Sophos; i="5.07,450,1413244800"; d="scan'208,217"; a="99624365" Received: from alln-core-11.cisco.com ([173.36.13.133]) by alln-iport-4.cisco.com with ESMTP; 24 Nov 2014 17:04:08 +0000 Received: from xhc-aln-x07.cisco.com (xhc-aln-x07.cisco.com [173.36.12.81]) by alln-core-11.cisco.com (8.14.5/8.14.5) with ESMTP id sAOH4836003517 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Mon, 24 Nov 2014 17:04:08 GMT Received: from xmb-rcd-x03.cisco.com ([169.254.7.228]) by xhc-aln-x07.cisco.com ([173.36.12.81]) with mapi id 14.03.0195.001; Mon, 24 Nov 2014 11:04:07 -0600 From: "Nagendra Kumar Nainar (naikumar)" To: "gjshep@gmail.com" , "bier@ietf.org" Thread-Topic: [Bier] Virtual Interim Meeting Thread-Index: AQHQCAL1vxphBtEE2kSOOt8oe8o9qZxwEguA Date: Mon, 24 Nov 2014 17:04:07 +0000 Message-ID: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.3.8.130913 x-originating-ip: [10.24.160.19] Content-Type: multipart/alternative; boundary="_000_D098CF131F4D2naikumarciscocom_" MIME-Version: 1.0 Archived-At: http://mailarchive.ietf.org/arch/msg/bier/uPpXgvXv9I2v3MbVXg21V4j3Tv0 Subject: Re: [Bier] Virtual Interim Meeting X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 24 Nov 2014 17:04:12 -0000 --_000_D098CF131F4D2naikumarciscocom_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi Greg, I am fine with the dates. Regards, Nagendra From: Greg Shepherd > Reply-To: "gjshep@gmail.com" > Date: Monday, November 24, 2014 at 11:22 AM To: "bier@ietf.org" > Subject: [Bier] Virtual Interim Meeting Thank all of you who participated in our first BIER BoF! I think the commun= ity response was overwhelmingly positive, but that just sets us up to get t= his work moving. I'd like to propose we have a virtual interim meeting to focus on the exist= ing work, and assign editor/contributors for each draft - getting commitmen= ts to both get the work progressing AND to show the ADs that we have commit= ted people to progress the work - and to review any new work that may not h= ave been presented at the BoF. I know this is a difficult part of the year, but let's see if we can find a= day/time that works for the majority. My proposed dates are: Dec 11, 12, or 15. I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I don'= t mind getting up at odd hours for a call if needed. Can each of the editors of the current drafts please speak up here as to wh= ich day/times work best for you. And anyone with new work that they wish to= present please respond as well. Looking forward to working with all of you. Cheers, Greg --_000_D098CF131F4D2naikumarciscocom_ Content-Type: text/html; charset="us-ascii" Content-ID: Content-Transfer-Encoding: quoted-printable
Hi Greg,

I am fine with the dates.

Regards,
Nagendra

From: Greg Shepherd <gjshep@gmail.com>
Reply-To: "gjshep@gmail.com" <gjshep@gmail.com>
Date: Monday, November 24, 2014 at = 11:22 AM
To: "bier@ietf.org" <bier@i= etf.org>
Subject: [Bier] Virtual Interim Mee= ting

Thank all of you who participated in our first BIER BoF! I= think the community response was overwhelmingly positive, but that just se= ts us up to get this work moving.

I'd like to propose we have a virtual interim meeting to focus on the = existing work, and assign editor/contributors for each draft - getting comm= itments to both get the work progressing AND to show the ADs that we have c= ommitted people to progress the work - and to review any new work that may not have been presented at the = BoF.

I know this is a difficult part of the year, but let's see if we can f= ind a day/time that works for the majority. My proposed dates are:

Dec 11, 12, or 15.

I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I= don't mind getting up at odd hours for a call if needed.

Can each of the editors of the current drafts please speak up here as = to which day/times work best for you. And anyone with new work that they wi= sh to present please respond as well.

Looking forward to working with all of you.

Cheers,
Greg
 --_000_D098CF131F4D2naikumarciscocom_-- From nobody Tue Nov 25 09:59:36 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2BBDE1A6EED for ; Tue, 25 Nov 2014 09:59:35 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.909 X-Spam-Level: X-Spam-Status: No, score=-6.909 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, T_RP_MATCHES_RCVD=-0.01] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id otDcNs_GdUCM for ; Tue, 25 Nov 2014 09:59:33 -0800 (PST) Received: from smtp-fr.alcatel-lucent.com (fr-hpida-esg-02.alcatel-lucent.com [135.245.210.21]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1285C1A1F02 for ; Tue, 25 Nov 2014 09:59:33 -0800 (PST) Received: from us70uusmtp4.zam.alcatel-lucent.com (unknown [135.5.2.66]) by Websense Email Security Gateway with ESMTPS id CB9E2F796E799; Tue, 25 Nov 2014 17:59:27 +0000 (GMT) Received: from US70UWXCHHUB01.zam.alcatel-lucent.com (us70uwxchhub01.zam.alcatel-lucent.com [135.5.2.48]) by us70uusmtp4.zam.alcatel-lucent.com (GMO) with ESMTP id sAPHxT8W001387 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 25 Nov 2014 12:59:29 -0500 Received: from US70UWXCHMBA03.zam.alcatel-lucent.com ([169.254.9.112]) by US70UWXCHHUB01.zam.alcatel-lucent.com ([135.5.2.48]) with mapi id 14.03.0195.001; Tue, 25 Nov 2014 12:59:29 -0500 From: "Dolganow, Andrew (Andrew)" To: "Nagendra Kumar Nainar (naikumar)" , "gjshep@gmail.com" , "bier@ietf.org" Thread-Topic: [Bier] Virtual Interim Meeting Thread-Index: AQHQCNmOF5Ic4Fw1tkmG2K+pwkhXWg== Date: Tue, 25 Nov 2014 17:59:28 +0000 Message-ID: References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.6.141106 x-originating-ip: [135.5.27.16] Content-Type: multipart/alternative; boundary="_000_D09A2D7262F12andrewdolganowalcatellucentcom_" MIME-Version: 1.0 Archived-At: http://mailarchive.ietf.org/arch/msg/bier/5ZKrPFJcEem1xP1S3B5-mg8ldF4 Subject: Re: [Bier] Virtual Interim Meeting X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Nov 2014 17:59:35 -0000 --_000_D09A2D7262F12andrewdolganowalcatellucentcom_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Greg, 11 and 12 are best, I am likely to be en-route on the 15th Andrew On 2014-11-24, 12:04 PM, "Nagendra Kumar Nainar (naikumar)" wrote: Hi Greg, I am fine with the dates. Regards, Nagendra From: Greg Shepherd > Reply-To: "gjshep@gmail.com" > Date: Monday, November 24, 2014 at 11:22 AM To: "bier@ietf.org" > Subject: [Bier] Virtual Interim Meeting Thank all of you who participated in our first BIER BoF! I think the commun= ity response was overwhelmingly positive, but that just sets us up to get t= his work moving. I'd like to propose we have a virtual interim meeting to foc us on the existing work, and assign editor/contributors for each draft - ge= tting commitments to both get the work progressing AND to show the ADs that= we have committed people to progress the work - and to review any new work= that may not have been presented at the BoF. I know this is a difficult part of the year, but let's see if we can find a= day/time that works for the majority. My proposed dates are: Dec 11, 12, or 15. I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I don'= t mind getting up at odd hours for a call if needed. Can each of the editors of the current drafts please speak up here as to wh= ich day/times work best for you. And anyone with new work that they wish to= present please respond as well. Looking forward to working with all of you. Cheers, Greg --_000_D09A2D7262F12andrewdolganowalcatellucentcom_ Content-Type: text/html; charset="us-ascii" Content-ID: <4AB700AB283A8449B7B0DAE008FA8786@exchange.lucent.com> Content-Transfer-Encoding: quoted-printable
Greg,

11 and 12 are best, I am likely to be en-route on the 15th 

Andrew


On 2014-11-24, 12:04 PM, "Nagendra Kumar Nainar (naikumar)" = wrote:

Hi Greg,

I am fine with the dates.

Regards,
Nagendra

From: Greg Shepherd <gjshep@gmail.com>
Reply-To: "gjshep@gmail.com" <gjshep@gmail.com>
Date: Monday, November 24, 2014 at = 11:22 AM
To: "bier@ietf.org" <bier@i= etf.org>
Subject: [Bier] Virtual Interim Mee= ting

Thank all of you who participated in our first BIER BoF! I= think the community response was overwhelmingly positive, but that just se= ts us up to get this work moving.

I'd like to propose we have a virtual interim meeting to foc
us on the existing work, and assign editor/contributors for each draft= - getting commitments to both get the work progressing AND to show the ADs= that we have committed people to progress the work - and to review any new= work that may not have been presented at the BoF.

I know this is a difficult part of the year, but let's see if we can f= ind a day/time that works for the majority. My proposed dates are:

Dec 11, 12, or 15.

I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I= don't mind getting up at odd hours for a call if needed.

Can each of the editors of the current drafts please speak up here as = to which day/times work best for you. And anyone with new work that they wi= sh to present please respond as well.

Looking forward to working with all of you.

Cheers,
Greg
 --_000_D09A2D7262F12andrewdolganowalcatellucentcom_-- From nobody Tue Nov 25 10:14:53 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 190D51A6F6D for ; Tue, 25 Nov 2014 10:14:48 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.2 X-Spam-Level: X-Spam-Status: No, score=-4.2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 2hJtbhCfoUeL for ; Tue, 25 Nov 2014 10:14:40 -0800 (PST) Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 67F811A6FD1 for ; Tue, 25 Nov 2014 10:14:39 -0800 (PST) X-AuditID: c6180641-f79916d00000623a-9d-54746b3964c6 Received: from EUSAAHC001.ericsson.se (Unknown_Domain [147.117.188.75]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id 9B.97.25146.93B64745; Tue, 25 Nov 2014 12:42:50 +0100 (CET) Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC001.ericsson.se ([147.117.188.75]) with mapi id 14.03.0195.001; Tue, 25 Nov 2014 13:14:31 -0500 From: Jeff Tantsura To: "gjshep@gmail.com" , "bier@ietf.org" Thread-Topic: [Bier] Virtual Interim Meeting Thread-Index: AQHQCAL1z3RyRbcNT06fy78LLAxAm5xxdPwA Date: Tue, 25 Nov 2014 18:14:30 +0000 Message-ID: References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.4.140807 x-originating-ip: [147.117.188.10] Content-Type: multipart/alternative; boundary="_000_D09A06DA7C298jefftantsuraericssoncom_" MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFmphkeLIzCtJLcpLzFFi42KZXLrHW9cquyTEYOcKSYulM/YwWWyZk+fA 5LFz1l12jyVLfjIFMEVx2aSk5mSWpRbp2yVwZazqucFa0KRR8aOpkbWB8aFiFyMnh4SAicTu XVOZIWwxiQv31rN1MXJxCAkcYZTobZnECOEsZ5Q4u/gXO0gVm4CBxP9vx1lAbBEBL4k9t2cx gdjCAtoSJ3d+Z4SI60h0vZ3JBGEbScz48xtsA4uAqsSf7zPYQGxeAXOJwxPeg80UEgiQ+H34 KFicUyBQYuff+2C9jEAXfT+1BsxmFhCXuPVkPhPEpQISS/ach7paVOLl43+sILaogJ7Esw2b 2SHiShKTlp5jheiNkfi/ex47xF5BiZMzn7BMYBSdhWTsLCRls5CUQcR1JBbs/sQGYWtLLFv4 mhnGPnPgMVSvtcTN5QuYkdUsYORYxchRWpxalptuZLiJERhtxyTYHHcwLvhkeYhRgINRiYfX oKE4RIg1say4MvcQozQHi5I4r2b1vGAhgfTEktTs1NSC1KL4otKc1OJDjEwcnFINjGGBqT/r veM2fjP6J8vlvLSwpezPIdfjMizeVafcfwVF7pol5prbxu+5KIxVc2LGZt2TU3eVTLz/+Yrd aS3Gh4xNas6thg1VzXEy823blycfKlT/qHNjs8/NJnf/ZWnJ99q72tN3xPay18dmzHyuWye+ iilZtz/L8o2AnaVY3mIG7ytaWpOVWIozEg21mIuKEwEQvX34lwIAAA== Archived-At: http://mailarchive.ietf.org/arch/msg/bier/6kJhrADrn59l4ljyb2SLsU96aHg Subject: Re: [Bier] Virtual Interim Meeting X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Nov 2014 18:14:49 -0000 --_000_D09A06DA7C298jefftantsuraericssoncom_ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Hi Greg, 15 would be the best. Thanks! Cheers, Jeff From: Greg Shepherd > Reply-To: "gjshep@gmail.com" > Date: Monday, November 24, 2014 at 8:22 AM To: "bier@ietf.org" > Subject: [Bier] Virtual Interim Meeting Thank all of you who participated in our first BIER BoF! I think the commun= ity response was overwhelmingly positive, but that just sets us up to get t= his work moving. I'd like to propose we have a virtual interim meeting to focus on the exist= ing work, and assign editor/contributors for each draft - getting commitmen= ts to both get the work progressing AND to show the ADs that we have commit= ted people to progress the work - and to review any new work that may not h= ave been presented at the BoF. I know this is a difficult part of the year, but let's see if we can find a= day/time that works for the majority. My proposed dates are: Dec 11, 12, or 15. I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I don'= t mind getting up at odd hours for a call if needed. Can each of the editors of the current drafts please speak up here as to wh= ich day/times work best for you. And anyone with new work that they wish to= present please respond as well. Looking forward to working with all of you. Cheers, Greg --_000_D09A06DA7C298jefftantsuraericssoncom_ Content-Type: text/html; charset="us-ascii" Content-ID: <10135F34D95A4243922BCA6A83A09975@ericsson.com> Content-Transfer-Encoding: quoted-printable
Hi Greg,

15 would be the best.
Thanks!

Cheers,
Jeff

From: Greg Shepherd <gjshep@gmail.com>
Reply-To: "gjshep@gmail.com" <gjshep@gmail.com>
Date: Monday, November 24, 2014 at = 8:22 AM
To: "bier@ietf.org" <bier@i= etf.org>
Subject: [Bier] Virtual Interim Mee= ting

Thank all of you who participated in our first BIER BoF! I= think the community response was overwhelmingly positive, but that just se= ts us up to get this work moving.

I'd like to propose we have a virtual interim meeting to focus on the = existing work, and assign editor/contributors for each draft - getting comm= itments to both get the work progressing AND to show the ADs that we have c= ommitted people to progress the work - and to review any new work that may not have been presented at the = BoF.

I know this is a difficult part of the year, but let's see if we can f= ind a day/time that works for the majority. My proposed dates are:

Dec 11, 12, or 15.

I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I= don't mind getting up at odd hours for a call if needed.

Can each of the editors of the current drafts please speak up here as = to which day/times work best for you. And anyone with new work that they wi= sh to present please respond as well.

Looking forward to working with all of you.

Cheers,
Greg
 --_000_D09A06DA7C298jefftantsuraericssoncom_-- From nobody Tue Nov 25 10:15:00 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5E2721A6F6D for ; Tue, 25 Nov 2014 10:14:50 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -14.511 X-Spam-Level: X-Spam-Status: No, score=-14.511 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, RCVD_IN_DNSWL_HI=-5, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01, USER_IN_DEF_DKIM_WL=-7.5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wUL8Oqv98Zpw for ; Tue, 25 Nov 2014 10:14:46 -0800 (PST) Received: from alln-iport-2.cisco.com (alln-iport-2.cisco.com [173.37.142.89]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 1221E1A6FDC for ; Tue, 25 Nov 2014 10:14:43 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=cisco.com; i=@cisco.com; l=2117; q=dns/txt; s=iport; t=1416939285; x=1418148885; h=from:to:subject:date:message-id:references:in-reply-to: content-id:content-transfer-encoding:mime-version; bh=BRH8jIlwN/9xKF8ygBKLIk3KlAIrTweNrYjh5m48tt0=; b=XEzaICpm6Eebp0FpkBKW5+0qSbrjrxJx409LB/XQi3aLov2Y79FH3kwI yfw/LV1HgZY81Z5aqLk/EYLIrNxyq9ctqxNVaYJrtloz6QdAuHsAeYjmy /ppQXPc2uGCkPaDmN74mUkXIxCt+XsX1AZ5Ej96A1zTv2Sr/6ZdrHPlQ4 8=; X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AiIFAELGdFStJV2a/2dsb2JhbABbgwaBKQTPcAKBExYBAQEBAX2EAgEBAQQ6SwQCAQgRAwECAR4QIREUCQgCBAESFIgXAxLLKA2GOAEBAQEBAQEBAgEBAQEBAQEBARmOQ4INMgaERwWSY4oFghSBNYNZixGGfYN8d4FIgQIBAQE X-IronPort-AV: E=Sophos;i="5.07,456,1413244800"; d="scan'208";a="100024456" Received: from rcdn-core-3.cisco.com ([173.37.93.154]) by alln-iport-2.cisco.com with ESMTP; 25 Nov 2014 18:14:44 +0000 Received: from xhc-aln-x08.cisco.com (xhc-aln-x08.cisco.com [173.36.12.82]) by rcdn-core-3.cisco.com (8.14.5/8.14.5) with ESMTP id sAPIEhpt013594 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 25 Nov 2014 18:14:43 GMT Received: from xmb-rcd-x06.cisco.com ([169.254.6.58]) by xhc-aln-x08.cisco.com ([173.36.12.82]) with mapi id 14.03.0195.001; Tue, 25 Nov 2014 12:14:42 -0600 From: "Rajiv Asati (rajiva)" To: "Dolganow, Andrew (Andrew)" , "Nagendra Kumar Nainar (naikumar)" , "gjshep@gmail.com" , "bier@ietf.org" Thread-Topic: [Bier] Virtual Interim Meeting Thread-Index: AQHQCAL1EbUiglWxMEqhszTVRRZSUJxxuBWA Date: Tue, 25 Nov 2014 18:14:42 +0000 Message-ID: References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: user-agent: Microsoft-MacOutlook/14.4.3.140616 x-originating-ip: [10.82.249.130] Content-Type: text/plain; charset="us-ascii" Content-ID: Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Archived-At: http://mailarchive.ietf.org/arch/msg/bier/BIzHqa88wiSXiW9KP_tH3jBQHmQ Subject: Re: [Bier] Virtual Interim Meeting X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Nov 2014 18:14:50 -0000 12th or 15th would work for me, but not the 11th. --=20 Cheers, Rajiv Asati Distinguished Engineer, Cisco -----Original Message----- From: , "Andrew (Andrew)" Date: Tuesday, November 25, 2014 at 12:59 PM To: Nagendra Kumar , "gjshep@gmail.com" , "bier@ietf.org" Subject: Re: [Bier] Virtual Interim Meeting >Greg, > > >11 and 12 are best, I am likely to be en-route on the 15th > > >Andrew > > > > >On 2014-11-24, 12:04 PM, "Nagendra Kumar Nainar (naikumar)" wrote: > > > >>Hi Greg, >> >> >>I am fine with the dates. >> >> >>Regards, >>Nagendra >> >> >>From: Greg Shepherd >>Reply-To: "gjshep@gmail.com" >>Date: Monday, November 24, 2014 at 11:22 AM >>To: "bier@ietf.org" >>Subject: [Bier] Virtual Interim Meeting >> >> >> >>Thank all of you who participated in our first BIER BoF! I think the >>community response was overwhelmingly positive, but that just sets us up >>to get this work moving. >> >> >>I'd like to propose we have a virtual interim meeting to foc >> >> >> >> >> >>us on the existing work, and assign editor/contributors for each draft - >>getting commitments to both get the work progressing AND to show the ADs >>that we have committed people to progress the work - and to review any >>new work that may not have been presented >> at the BoF. >> >> >>I know this is a difficult part of the year, but let's see if we can >>find a day/time that works for the majority. My proposed dates are: >> >> >>Dec 11, 12, or 15. >> >> >>I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I >>don't mind getting up at odd hours for a call if needed. >> >> >>Can each of the editors of the current drafts please speak up here as to >>which day/times work best for you. And anyone with new work that they >>wish to present please respond as well. >> >> >>Looking forward to working with all of you. >> >> >>Cheers, >>Greg >> >> >> >> >> From nobody Tue Nov 25 10:36:20 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9E6411A8777 for ; Tue, 25 Nov 2014 10:35:05 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2 X-Spam-Level: X-Spam-Status: No, score=-2 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NkXTRDYkPWT6 for ; Tue, 25 Nov 2014 10:35:03 -0800 (PST) Received: from mail-pd0-x232.google.com (mail-pd0-x232.google.com [IPv6:2607:f8b0:400e:c02::232]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 46EC41A8763 for ; Tue, 25 Nov 2014 10:35:03 -0800 (PST) Received: by mail-pd0-f178.google.com with SMTP id g10so1088387pdj.9 for ; Tue, 25 Nov 2014 10:35:02 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=content-type:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=tbBxScVZGj3eRuRyRDdkFkWoa767Z/orL7MUKd9j7JA=; b=KmdqwLJh00k6JZcq+2GO4AH+bKMASH+cOPsehe/1JVWcDk06zoEQdzlGutmzTdkHh8 rNlWg7T91xcc7LFIcq/N65ik9xehAT1dDhm1M5ICqSuzAdmZU+JHkBs75wuvh05HAxVw 5SsE3tSnmU5WdY27UpDczHgOG2anmKcNlr58t6Hk8yPNYQycdEBuXiSeqFCLik2K4B4Q dnxq34kMwpvQFCqKV9Da/KtrkW2wkvssUxYhc0rb+zLDukOKIRPXjf8zNLhyFH2LVxWi jJ+q6JE/mwqJIUr9UEcMfljub6lJ53n/+sD7FUx+bnAhzcLtEctfGq/7fFzfrv2+nv4T MXVQ== X-Received: by 10.70.135.41 with SMTP id pp9mr46033079pdb.77.1416940502563; Tue, 25 Nov 2014 10:35:02 -0800 (PST) Received: from [10.74.94.72] ([166.170.38.237]) by mx.google.com with ESMTPSA id a6sm2121878pbu.64.2014.11.25.10.34.59 for (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Tue, 25 Nov 2014 10:35:01 -0800 (PST) Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (1.0) From: Sam Aldrin X-Mailer: iPhone Mail (12B436) In-Reply-To: Date: Tue, 25 Nov 2014 10:34:58 -0800 Content-Transfer-Encoding: quoted-printable Message-Id: <165A1E0D-B403-4344-961D-8D2EA6FC00EB@gmail.com> References: To: "Rajiv Asati (rajiva)" Archived-At: http://mailarchive.ietf.org/arch/msg/bier/rsFk65iHbu4GaDasGLmXrL0v3vE Cc: "bier@ietf.org" , "Nagendra Kumar Nainar \(naikumar\)" , "Dolganow, Andrew \(Andrew\)" , "gjshep@gmail.com" Subject: Re: [Bier] Virtual Interim Meeting X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Nov 2014 18:35:05 -0000 11th or 12th works for me. -sam Sent from my iPhone > On Nov 25, 2014, at 10:14 AM, Rajiv Asati (rajiva) wrot= e: >=20 > 12th or 15th would work for me, but not the 11th. >=20 > --=20 > Cheers, > Rajiv Asati > Distinguished Engineer, Cisco >=20 >=20 >=20 >=20 >=20 > -----Original Message----- > From: , "Andrew (Andrew)" = > Date: Tuesday, November 25, 2014 at 12:59 PM > To: Nagendra Kumar , "gjshep@gmail.com" > , "bier@ietf.org" > Subject: Re: [Bier] Virtual Interim Meeting >=20 >> Greg, >>=20 >>=20 >> 11 and 12 are best, I am likely to be en-route on the 15th >>=20 >>=20 >> Andrew >>=20 >>=20 >>=20 >>=20 >> On 2014-11-24, 12:04 PM, "Nagendra Kumar Nainar (naikumar)" wrote: >>=20 >>=20 >>=20 >>> Hi Greg, >>>=20 >>>=20 >>> I am fine with the dates. >>>=20 >>>=20 >>> Regards, >>> Nagendra >>>=20 >>>=20 >>> From: Greg Shepherd >>> Reply-To: "gjshep@gmail.com" >>> Date: Monday, November 24, 2014 at 11:22 AM >>> To: "bier@ietf.org" >>> Subject: [Bier] Virtual Interim Meeting >>>=20 >>>=20 >>>=20 >>> Thank all of you who participated in our first BIER BoF! I think the >>> community response was overwhelmingly positive, but that just sets us up= >>> to get this work moving. >>>=20 >>>=20 >>> I'd like to propose we have a virtual interim meeting to foc >>>=20 >>>=20 >>>=20 >>>=20 >>>=20 >>> us on the existing work, and assign editor/contributors for each draft -= >>> getting commitments to both get the work progressing AND to show the ADs= >>> that we have committed people to progress the work - and to review any >>> new work that may not have been presented >>> at the BoF. >>>=20 >>>=20 >>> I know this is a difficult part of the year, but let's see if we can >>> find a day/time that works for the majority. My proposed dates are: >>>=20 >>>=20 >>> Dec 11, 12, or 15. >>>=20 >>>=20 >>> I'll be in EU on the 11th and 12th, and back in PDT on the 15th. But I >>> don't mind getting up at odd hours for a call if needed. >>>=20 >>>=20 >>> Can each of the editors of the current drafts please speak up here as to= >>> which day/times work best for you. And anyone with new work that they >>> wish to present please respond as well. >>>=20 >>>=20 >>> Looking forward to working with all of you. >>>=20 >>>=20 >>> Cheers, >>> Greg >=20 > _______________________________________________ > BIER mailing list > BIER@ietf.org > https://www.ietf.org/mailman/listinfo/bier From nobody Wed Nov 26 02:06:15 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 143DC1A89E1 for ; Wed, 26 Nov 2014 02:06:14 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -4.211 X-Spam-Level: X-Spam-Status: No, score=-4.211 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id uLeb-zXjHfgU for ; Wed, 26 Nov 2014 02:06:11 -0800 (PST) Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 9D6A21A89EF for ; Wed, 26 Nov 2014 02:06:10 -0800 (PST) Received: from 172.18.7.190 (EHLO lhreml406-hub.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.7-GA FastPath queued) with ESMTP id BPG02430; Wed, 26 Nov 2014 10:06:09 +0000 (GMT) Received: from NKGEML403-HUB.china.huawei.com (10.98.56.34) by lhreml406-hub.china.huawei.com (10.201.5.243) with Microsoft SMTP Server (TLS) id 14.3.158.1; Wed, 26 Nov 2014 10:06:01 +0000 Received: from NKGEML512-MBS.china.huawei.com ([169.254.8.22]) by nkgeml403-hub.china.huawei.com ([10.98.56.34]) with mapi id 14.03.0158.001; Wed, 26 Nov 2014 18:05:50 +0800 From: Xuxiaohu To: Albert Tian , AshwoodsmithPeter , "bier@ietf.org" Thread-Topic: [Bier] BIER and deterministic ECMP Thread-Index: AQHQAvPmimRcG0ie5kmRBDyo9YL80ZxmvLFw///jxoCADBoxsA== Date: Wed, 26 Nov 2014 10:05:50 +0000 Message-ID: <1FEE3F8F5CCDE64C9A8E8F4AD27C19EE082D8373@NKGEML512-MBS.china.huawei.com> References: <7AE6A4247B044C4ABE0A5B6BF427F8E20A43D4EE@dfweml703-chm> In-Reply-To: Accept-Language: zh-CN, en-US Content-Language: zh-CN X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.111.98.134] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-CFilter-Loop: Reflected Archived-At: http://mailarchive.ietf.org/arch/msg/bier/h1jICHYvh-mbtGbRsZ8R1fPqbz8 Subject: Re: [Bier] BIER and deterministic ECMP X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Nov 2014 10:06:14 -0000 For deterministic path selection for each egress, how about disabling the E= CMP for BIER forwarding? Best regards, Xiaohu > -----Original Message----- > From: BIER [mailto:bier-bounces@ietf.org] On Behalf Of Albert Tian > Sent: Wednesday, November 19, 2014 4:15 AM > To: AshwoodsmithPeter; bier@ietf.org > Subject: Re: [Bier] BIER and deterministic ECMP >=20 > Hi Peter, >=20 > If I understood it correctly, your proposal is still non-deterministic. > Consider fluctuation of group members behind the lower ordered links, the= y > could still impact the paths of the members behind the higher ordered lin= ks, > assuming you are going to have F-BM on for each link. >=20 > Statistically speaking, lower ordered links would also draw more traffic = due to > the F-BM operations. >=20 > Thanks, > Albert >=20 >=20 >=20 >=20 >=20 > On 11/18/14, 11:00 AM, "AshwoodsmithPeter" > > wrote: >=20 > >What if rather than process in the order of the bits the processing is > >inverted and instead the links are processed in some lexicographic > >ordering as the outer loop. That would lead to a determinism of the > >paths used irrespective of entry or departure of lower order bits. > >Slower I guess for a sparsely populated bit set. > > > >Peter > > > >-----Original Message----- > >From: BIER [mailto:bier-bounces@ietf.org] On Behalf Of Albert Tian > >Sent: Tuesday, November 18, 2014 12:53 AM > >To: Ross Callon; bier@ietf.org > >Subject: Re: [Bier] BIER and deterministic ECMP > > > >Hi Ross, > > > >Very good point. Yes IMHO this needs to be addressed, otherwise the > >links leading to lower numbered BFERs will more likely be chosen in > >ECMP and cause unbalanced traffic. This maybe just as bad as the > >instability you have identified. > > > >Agreed multiple BIFT solution should work. > > > >Thanks, > >Albert > > > > > > > > > > > >On 11/17/14, 3:10 PM, "Ross Callon" wrote: > > > >>Reading the BIER architecture (draft-wijnands-bier-architecture-01) I > >>noticed the last sentence of section 6.7: > >> > >> Note however that by the rules of Section 6.5, any packet destined > >> for both BFER-D and BFER-F will be sent via BFR-C. > >> > >>Thinking about this more: It is clear that where ECMP paths exist, > >>which path is taken to reach any particular destination will depend > >>upon which other destinations the same packet goes to. In particular, > >>which lower numbered bits are set in the BitString can change the > >>choice among ECMP paths. This could have the effect that as a lower > >>numbered destination intermittently joins and leaves a multicast > >>group, the path to a higher numbered destination could switch back and > >>forth between different ECMP paths. > >> > >>This could be fixed if we used multiple BIFT tables, such that each > >>BIFT table has only one path to each destination, but the multiple > >>possible ECMP paths to any particular destination are spread across > >>the multiple tables. When a BIER-encapsulated packet arrives to be > >>forwarded, the BFR would use the BIER Entropy field (or a hash > >>thereon, if fewer than 256 next hops are > >>supported) > >>to determine which BIFT to use, and then the normal BIER forwarding > >>algorithm (the same as described in > >>draft-wijnands-bier-architecture-01) would be used with whichever BIFT > >>table was selected. This would of course increase the amount of memory > >>that would be needed to store BIFT tables, but my quick "back of the > >>envelope" calculation suggests that the amount of memory to store BIFT > >>tables still should not be a problem if there aren't too many BIFT > >>tables (I did my "back of the envelope" with max 16 tables). The > >>forwarding algorithm and packet headers stay unchanged. This also has > >>the feature that if this approach is consistently implemented across a > >>service provider, then where a problem has been reported on a > >>particular flow of packets to a particular receiver, we can traceroute > >>with the same entropy value to that particular receiver and know that > >>we will be using the same path as the normal data packets to that > >>receiver. > >> > >>I would be happy to write up appropriate text to add to > >>draft-wijnands-bier- > >>architecture-01 (presumably for a -02 version) if the authors would > >>like, or alternately could write up a separate Internet Draft to > >>further describe the issue and the solution. > >> > >>Thanks, Ross > >> > >> > >> > >>_______________________________________________ > >>BIER mailing list > >>BIER@ietf.org > >>https://www.ietf.org/mailman/listinfo/bier > > > >_______________________________________________ > >BIER mailing list > >BIER@ietf.org > >https://www.ietf.org/mailman/listinfo/bier > > > >_______________________________________________ > >BIER mailing list > >BIER@ietf.org > >https://www.ietf.org/mailman/listinfo/bier >=20 > _______________________________________________ > BIER mailing list > BIER@ietf.org > https://www.ietf.org/mailman/listinfo/bier From nobody Fri Nov 28 10:54:09 2014 Return-Path: X-Original-To: bier@ietfa.amsl.com Delivered-To: bier@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BB96E1A0113; Fri, 28 Nov 2014 10:54:07 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.999 X-Spam-Level: X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id YXy6JRwF0i7O; Fri, 28 Nov 2014 10:54:06 -0800 (PST) Received: from mail-wi0-x22f.google.com (mail-wi0-x22f.google.com [IPv6:2a00:1450:400c:c05::22f]) (using TLSv1 with cipher ECDHE-RSA-RC4-SHA (128/128 bits)) (No client certificate requested) by ietfa.amsl.com (Postfix) with ESMTPS id 56F101A010C; Fri, 28 Nov 2014 10:54:06 -0800 (PST) Received: by mail-wi0-f175.google.com with SMTP id l15so19372901wiw.14 for ; Fri, 28 Nov 2014 10:54:05 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:reply-to:date:message-id:subject:from:to:content-type; bh=7vHrTmGYlLmaOl1c+hLYEETEd8KN5w9sfL/c0D1pNYw=; b=bhYAKeNT1JB1GAUs9nRu0dWkiSFqiYAdHQx/odliPSEHideIaneGP2J0O+wdBUEcRj yQutCBPVy80FExkODWyACfxjknTDqkpmWR9kwwc0zy0kiUmsObYlsme+42973oDzwEDz Msp+8Ha6VWE2lOCgUtRLRkE1k1aWGg5/7MMDmcWGxVHs2gIh3oBYNvoyqp7Lx0n+pBM/ XB9GNOk5RJmiNqVPOZ2hs+qUvBVc+uX31YnFkS10Xr1e6yBp1kbiEC7FW69TMVsNYoWa DP/rJU6FjpYFWdzy3M7qnJWj2/LKExl4F6OStN6b1dk6dcbTr608EhC8/p4YNo7zdmBQ HQog== MIME-Version: 1.0 X-Received: by 10.180.186.40 with SMTP id fh8mr63465019wic.40.1417200845146; Fri, 28 Nov 2014 10:54:05 -0800 (PST) Received: by 10.180.207.200 with HTTP; Fri, 28 Nov 2014 10:54:05 -0800 (PST) Date: Fri, 28 Nov 2014 10:54:05 -0800 Message-ID: From: Greg Shepherd To: "iesg-secretary@ietf.org" , "bier@ietf.org" Content-Type: multipart/alternative; boundary=001a11c2681cf34ffd0508efc86f Archived-At: http://mailarchive.ietf.org/arch/msg/bier/fQrJdmOeP_saH-rjruF-Rd5Kq2M Subject: [Bier] Interim Virtual Meeting Announcement X-BeenThere: bier@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gjshep@gmail.com List-Id: "\"Bit Indexed Explicit Replication discussion list\"" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Nov 2014 18:54:07 -0000 --001a11c2681cf34ffd0508efc86f Content-Type: text/plain; charset=UTF-8 BIER Virtual Interim Agenda Monday, Dec 15, 2014 0800-1000 PDT Welcome, note well, agenda bash, Chairs - 5min Confirmation of work items - 30min all current drafts any new proposed drafts Confirmation of editor+contributors to all confirmed work - 30min w/ target date of next submission Use Cases draft needs a wide range of contributors Potential Charter discussion - 30min WebEx Link TBA If you are an editor/contributor of an existing draft and CANNOT make this Interem Virtual meeting, please speak up to confirm your commitment (or not) to progress this work BEFORE the meeting so that we may take appropriate action. Thanks, Greg --001a11c2681cf34ffd0508efc86f Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable
BIER Virtual Interim Agenda
Monday, Dec 15,= 2014 0800-1000 PDT

Welcome, note well, agenda bash, Chairs - 5min
Confirmation of work = items - 30min
all current drafts
= any new proposed drafts
Confirmation of editor+contributors to all confirmed work = - 30min
w/ ta= rget date of next submission
Use Cases draft needs a wide range of contributors
Potential Charter disc= ussion - 30min
WebEx Link T= BA

If you are an editor/contributor of an existing draft and CANNOT make t= his Interem Virtual meeting, please speak up to confirm your commitment (or= not) to progress this work BEFORE the meeting so that we may take appropri= ate action.=C2=A0

Thanks,
Greg
--001a11c2681cf34ffd0508efc86f--