From jana.iyengar@gmail.com Sun Oct 9 16:51:48 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A7DB521F8B5C for ; Sun, 9 Oct 2011 16:51:48 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.599 X-Spam-Level: X-Spam-Status: No, score=-3.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6HpbmMf51KGX for ; Sun, 9 Oct 2011 16:51:48 -0700 (PDT) Received: from mail-yw0-f44.google.com (mail-yw0-f44.google.com [209.85.213.44]) by ietfa.amsl.com (Postfix) with ESMTP id E05E521F8B57 for ; Sun, 9 Oct 2011 16:51:47 -0700 (PDT) Received: by ywm3 with SMTP id 3so6080130ywm.31 for ; Sun, 09 Oct 2011 16:51:47 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=message-id:date:from:reply-to:organization:user-agent:mime-version :to:cc:subject:content-type:content-transfer-encoding; bh=jMT7imUx9f/KTQVD2AKU65lpO9SWIkH6kk5ibBvM/0U=; b=kWmxFNW9m81OqpHlRVFY6vjA+77PODTM9gZduuo93IkifXrft6I1wEfCsh3ZmoFx7k 1cbjce6Wpcg1GWpH56v9bZcs9T/tEwu0YlQnCANcoI2F/tQ+FbL2nDMLBekllGv6M+uj ofbEmLWkqloBR1AljIaGSQ598P9/FJX0GhDlI= Received: by 10.68.13.40 with SMTP id e8mr32560406pbc.111.1318204307169; Sun, 09 Oct 2011 16:51:47 -0700 (PDT) Received: from surutti.fandm.edu (c-76-103-130-241.hsd1.ca.comcast.net. [76.103.130.241]) by mx.google.com with ESMTPS id p4sm61078753pbs.6.2011.10.09.16.51.45 (version=TLSv1/SSLv3 cipher=OTHER); Sun, 09 Oct 2011 16:51:46 -0700 (PDT) Message-ID: <4E92338F.1030601@fandm.edu> Date: Sun, 09 Oct 2011 16:51:43 -0700 From: Janardhan Iyengar Organization: Franklin & Marshall College User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en-US; rv:1.9.2.23) Gecko/20110920 Lightning/1.0b2 Thunderbird/3.1.15 MIME-Version: 1.0 To: ledbat@ietf.org Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: janardhan.iyengar@fandm.edu List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 09 Oct 2011 23:51:48 -0000 Dear all, A new version of the congestion control draft is in the repository. There are two major mods in this revision: 1/ WGLC identified one major issue that needed to be addressed in the LEDBAT congestion control draft -- LEDBAT response to extreme congestion -- and we've tried to address that issue in this revision. We've added a new mechanism, the Congestion Timeout (CTO), for a sender to respond to extreme congestion. We do not specify how this should be implemented, but we do note that a CTO can be implemented with or without a timer. In terms of textual changes, we've added an update_CTO() function and a branch for what to do if no acks are received within a CTO amount of time in Section 3.4.2. We have changed the response to data loss to ensure that a protocol, such as TCP, that uses the same timer for both congestion control and for retransmissions, changes its cwnd correctly. 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and have clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in Section 3.5. Please comment! - jana -- Janardhan Iyengar Assistant Professor, Computer Science Franklin & Marshall College http://www.fandm.edu/jiyengar -------- Original Message -------- Subject: New Version Notification for draft-ietf-ledbat-congestion-08.txt Date: Sun, 09 Oct 2011 16:33:33 -0700 From: internet-drafts@ietf.org To: jiyengar@fandm.edu CC: jiyengar@fandm.edu, mirja.kuehlewind@ikr.uni-stuttgart.de, greg@bittorrent.com, shalunov@bittorrent.com A new version of I-D, draft-ietf-ledbat-congestion-08.txt has been successfully submitted by Janardhan Iyengar and posted to the IETF repository. Filename: draft-ietf-ledbat-congestion Revision: 08 Title: Low Extra Delay Background Transport (LEDBAT) Creation date: 2011-10-09 WG ID: ledbat Number of pages: 19 Abstract: LEDBAT is an experimental delay-based congestion control algorithm that attempts to utilize the available bandwidth on an end-to-end path while limiting the consequent increase in queueing delay on the path. LEDBAT uses changes in one-way delay measurements to limit congestion that the flow itself induces in the network. LEDBAT is designed for use by background bulk-transfer applications; it is designed to be no more aggressive than TCP congestion control and to yield in the presence of any competing flows when latency builds, thus limiting interference with the network performance of the competing flows. The IETF Secretariat From touch@isi.edu Mon Oct 10 07:38:18 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 69C1921F8BC3 for ; Mon, 10 Oct 2011 07:38:18 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.599 X-Spam-Level: X-Spam-Status: No, score=-102.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wHdL0rZBe6I6 for ; Mon, 10 Oct 2011 07:38:18 -0700 (PDT) Received: from darkstar.isi.edu (darkstar.isi.edu [128.9.128.127]) by ietfa.amsl.com (Postfix) with ESMTP id 0689421F8B85 for ; Mon, 10 Oct 2011 07:38:18 -0700 (PDT) Received: from [192.168.1.90] (pool-71-105-89-105.lsanca.dsl-w.verizon.net [71.105.89.105]) (authenticated bits=0) by darkstar.isi.edu (8.13.8/8.13.8) with ESMTP id p9AEbWAd024739 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Mon, 10 Oct 2011 07:37:37 -0700 (PDT) Message-ID: <4E93032B.7090006@isi.edu> Date: Mon, 10 Oct 2011 07:37:31 -0700 From: Joe Touch User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:6.0.2) Gecko/20110902 Thunderbird/6.0.2 MIME-Version: 1.0 To: janardhan.iyengar@fandm.edu References: <4E92338F.1030601@fandm.edu> In-Reply-To: <4E92338F.1030601@fandm.edu> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: Janardhan Iyengar , ledbat@ietf.org Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Oct 2011 14:38:18 -0000 On 10/9/2011 4:51 PM, Janardhan Iyengar wrote: > Dear all, > > A new version of the congestion control draft is in the repository. > There are two major mods in this revision: ... > 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and have > clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in > Section 3.5. It might be more stable to refer to TCP's INIT_CWND and MIN_CWND values, i.e., that ledbat should use TCP's values for these. (they've been under discussion for revision in TCPM, FWIW) Joe From mirja.kuehlewind@ikr.uni-stuttgart.de Mon Oct 10 09:46:30 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 154C021F8AE6 for ; Mon, 10 Oct 2011 09:46:30 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.249 X-Spam-Level: X-Spam-Status: No, score=-2.249 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HELO_EQ_DE=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UORwrxAE1qvN for ; Mon, 10 Oct 2011 09:46:29 -0700 (PDT) Received: from mailsrv.ikr.uni-stuttgart.de (mailsrv.ikr.uni-stuttgart.de [129.69.170.2]) by ietfa.amsl.com (Postfix) with ESMTP id 2BD7721F8A95 for ; Mon, 10 Oct 2011 09:46:29 -0700 (PDT) Received: from netsrv1.ikr.uni-stuttgart.de (netsrv1-c [10.11.12.12]) by mailsrv.ikr.uni-stuttgart.de (Postfix) with ESMTP id B98EF633B1; Mon, 10 Oct 2011 18:46:26 +0200 (CEST) Received: from vpn-2-cl177 (vpn-2-cl177 [10.41.21.177]) by netsrv1.ikr.uni-stuttgart.de (Postfix) with ESMTP id AA46159A8A; Mon, 10 Oct 2011 18:46:26 +0200 (CEST) From: Mirja Kuehlewind Organization: University of Stuttgart (Germany), IKR To: ledbat@ietf.org Date: Mon, 10 Oct 2011 18:46:26 +0200 User-Agent: KMail/1.9.10 (enterprise35 0.20101217.1207316) References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> In-Reply-To: <4E93032B.7090006@isi.edu> X-KMail-QuotePrefix: > MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Message-Id: <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> Cc: Janardhan Iyengar Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Oct 2011 16:46:30 -0000 Hi, I wouldn't want to refer to TCP's value because of a less-than-best-effort= =20 traffic there might always be a motivation to take different/lower values=20 than TCP. Mirja On Monday 10 October 2011 16:37:31 Joe Touch wrote: > On 10/9/2011 4:51 PM, Janardhan Iyengar wrote: > > Dear all, > > > > A new version of the congestion control draft is in the repository. > > There are two major mods in this revision: > > ... > > > 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and have > > clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in > > Section 3.5. > > It might be more stable to refer to TCP's INIT_CWND and MIN_CWND values, > i.e., that ledbat should use TCP's values for these. > > (they've been under discussion for revision in TCPM, FWIW) > > Joe > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat =2D-=20 =2D------------------------------------------------------------------ Dipl.-Ing. Mirja K=FChlewind Institute of Communication Networks and Computer Engineering (IKR) University of Stuttgart, Germany Pfaffenwaldring 47, D-70569 Stuttgart tel: +49(0)711/685-67973 email: mirja.kuehlewind@ikr.uni-stuttgart.de web: www.ikr.uni-stuttgart.de =2D------------------------------------------------------------------ From touch@isi.edu Mon Oct 10 10:46:48 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5610221F8C0C for ; Mon, 10 Oct 2011 10:46:48 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -101.999 X-Spam-Level: X-Spam-Status: No, score=-101.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_43=0.6, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4PvU2qmtERvK for ; Mon, 10 Oct 2011 10:46:48 -0700 (PDT) Received: from vapor.isi.edu (vapor.isi.edu [128.9.64.64]) by ietfa.amsl.com (Postfix) with ESMTP id E425F21F8BFE for ; Mon, 10 Oct 2011 10:46:47 -0700 (PDT) Received: from [128.9.160.166] (abc.isi.edu [128.9.160.166]) (authenticated bits=0) by vapor.isi.edu (8.13.8/8.13.8) with ESMTP id p9AHkVIg027006 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Mon, 10 Oct 2011 10:46:34 -0700 (PDT) Message-ID: <4E932F77.5060006@isi.edu> Date: Mon, 10 Oct 2011 10:46:31 -0700 From: Joe Touch User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1 MIME-Version: 1.0 To: Mirja Kuehlewind References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> In-Reply-To: <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: Janardhan Iyengar , ledbat@ietf.org Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Oct 2011 17:46:48 -0000 On 10/10/2011 9:46 AM, Mirja Kuehlewind wrote: > Hi, > > I wouldn't want to refer to TCP's value because of a less-than-best-effort > traffic there might always be a motivation to take different/lower values > than TCP. The values currently suggested are TCP's current values. I don't mind if you want to pick smaller ones, but I'd suggest spec'ing them relative to TCP's current values regardless. Joe > > Mirja > > > On Monday 10 October 2011 16:37:31 Joe Touch wrote: >> On 10/9/2011 4:51 PM, Janardhan Iyengar wrote: >>> Dear all, >>> >>> A new version of the congestion control draft is in the repository. >>> There are two major mods in this revision: >> >> ... >> >>> 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and have >>> clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in >>> Section 3.5. >> >> It might be more stable to refer to TCP's INIT_CWND and MIN_CWND values, >> i.e., that ledbat should use TCP's values for these. >> >> (they've been under discussion for revision in TCPM, FWIW) >> >> Joe >> _______________________________________________ >> ledbat mailing list >> ledbat@ietf.org >> https://www.ietf.org/mailman/listinfo/ledbat > > > From Rolf.Winter@neclab.eu Tue Oct 11 00:28:22 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 96C6C21F8D09 for ; Tue, 11 Oct 2011 00:28:22 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -99.999 X-Spam-Level: X-Spam-Status: No, score=-99.999 tagged_above=-999 required=5 tests=[BAYES_50=0.001, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9SQohmANi4Kw for ; Tue, 11 Oct 2011 00:28:22 -0700 (PDT) Received: from mailer1.neclab.eu (mailer1.neclab.eu [195.37.70.40]) by ietfa.amsl.com (Postfix) with ESMTP id 1A2E421F8D0C for ; Tue, 11 Oct 2011 00:28:21 -0700 (PDT) Received: from localhost (localhost.localdomain [127.0.0.1]) by mailer1.neclab.eu (Postfix) with ESMTP id E9438280001AA for ; Tue, 11 Oct 2011 09:28:18 +0200 (CEST) X-Virus-Scanned: Amavisd on Debian GNU/Linux (atlas1.office.hd) Received: from mailer1.neclab.eu ([127.0.0.1]) by localhost (atlas1.office.hd [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eml4qHda5oRJ for ; Tue, 11 Oct 2011 09:28:18 +0200 (CEST) Received: from METHONE.office.hd (Methone.office.hd [192.168.24.54]) by mailer1.neclab.eu (Postfix) with ESMTP id CF453280001A7 for ; Tue, 11 Oct 2011 09:28:13 +0200 (CEST) Received: from DAPHNIS.office.hd ([169.254.2.240]) by METHONE.office.hd ([192.168.24.54]) with mapi id 14.01.0323.003; Tue, 11 Oct 2011 09:27:42 +0200 From: Rolf Winter To: "ledbat@ietf.org" Thread-Topic: WG last call on draft-ietf-ledbat-congestion-08 Thread-Index: AcyH50J2tVNuoIWJTvi4Z4uw3dIKqw== Date: Tue, 11 Oct 2011 07:27:42 +0000 Message-ID: <791AD3077F94194BB2BDD13565B6295D1D0A463B@DAPHNIS.office.hd> Accept-Language: en-US, de-DE Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.7.0.206] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: [ledbat] WG last call on draft-ietf-ledbat-congestion-08 X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 11 Oct 2011 07:28:22 -0000 WG, this is to start a two week working group last call on draft-ietf-ledbat-co= ngestion-08 which will end 25th of October. Please review the document care= fully to make sure that previous comments have been addressed. In case you = think the document is ready to move forward, please state this on the list = and not just silently agree. Best, The chairs NEC Europe Limited | Registered Office: NEC House, 1 Victoria Road, London = W3 6BL | Registered in England 2832014=20 From bhooma@apple.com Tue Oct 11 16:40:00 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 64AC121F8B9C for ; Tue, 11 Oct 2011 16:40:00 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.599 X-Spam-Level: X-Spam-Status: No, score=-106.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5gIF4HYL+f6I for ; Tue, 11 Oct 2011 16:40:00 -0700 (PDT) Received: from mail-out.apple.com (mail-out.apple.com [17.151.62.51]) by ietfa.amsl.com (Postfix) with ESMTP id 0831921F8B66 for ; Tue, 11 Oct 2011 16:40:00 -0700 (PDT) MIME-version: 1.0 Content-transfer-encoding: 7BIT Content-type: text/plain; CHARSET=US-ASCII Received: from relay13.apple.com ([17.128.113.29]) by mail-out.apple.com (Oracle Communications Messaging Exchange Server 7u4-20.01 64bit (built Nov 21 2010)) with ESMTPS id <0LSX00DXADQKUBD1@mail-out.apple.com> for ledbat@ietf.org; Tue, 11 Oct 2011 16:39:59 -0700 (PDT) X-AuditID: 1180711d-b7b16ae000004efb-2a-4e94d2f8eafa Received: from [17.193.15.22] (Unknown_Domain [17.193.15.22]) by relay13.apple.com (Apple SCV relay) with SMTP id 60.5B.20219.8F2D49E4; Tue, 11 Oct 2011 16:36:24 -0700 (PDT) From: Padma Bhooma Date: Tue, 11 Oct 2011 16:39:59 -0700 Message-id: <2D481290-12BA-4101-81A1-07FF95A7CD29@apple.com> To: ledbat@ietf.org X-Mailer: Apple Mail (2.1247) X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFvrCJMWRmVeSWpSXmKPExsUieJBfTPfHpSl+Bs3X+C3+PXrO5sDosWTJ T6YAxigum5TUnMyy1CJ9uwSujPu7kguWsFZMfVPfwDiDpYuRk0NCwERias9ONghbTOLCvfVA NheHkMB6Rome9h9MIAk2AVWJk3/uMIPYzAJaEjf+vWSCsLUlli18DRYXFrCVOHb2MNhQFqD6 Z6c6wYbyCthIXHxyAywuIiAsMePUOVaIuKHEz8nrmCAWy0q8e/uGbQIjzywkK2YhWTELScsC RuZVjIJFqTmJlYbGeokFBTmpesn5uZsYQWHRUCi7g3H/T/5DjAIcjEo8vBu7J/sJsSaWFVfm HmKU4GBWEuHt3zXFT4g3JbGyKrUoP76oNCe1+BCjNAeLkjjvtTNAKYH0xJLU7NTUgtQimCwT B6dUA+OUvS+Fjj/pclNRnNYeHywQL5/GV5dcm5bn8PPGrVDOmiUHJ/nOq9ITy9z043vM1cu9 3CLxa3USNASCW258kw/RZ2/T0frncqivREZ0qRuPxdkH7EkVizzPP3lbWvuG5cXje1tlPqy9 Isw445DjVJNL6d9nTzCdd4qlgfVK4KZPj/dfL1yx5JISS3FGoqEWc1FxIgDWtfIfBwIAAA== Subject: Re: [ledbat] WG last call on draft-ietf-ledbat-congestion-08 X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 11 Oct 2011 23:40:00 -0000 Hi, I reviewed the latest draft and I have few comments on the latest changes. A receiver may have to echo the remote timestamp to allow for computation of RTT which it can use for computing CTO. It might be useful to mention that the sender is supposed to retransmit sent data or transmit new data to restart the ACK clock. Other than the comments above I think the algorithm is ready for adoption. The implementation of LEDBAT algorithm that was done as a TCP Congestion control module in Mac OS X is available at the following link: http://opensource.apple.com/source/xnu/xnu-1699.22.81/bsd/netinet/tcp_ledbat.c Thanks, Padma From Rolf.Winter@neclab.eu Fri Oct 14 11:07:47 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 230CC21F8CB4 for ; Fri, 14 Oct 2011 11:07:47 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.274 X-Spam-Level: X-Spam-Status: No, score=-102.274 tagged_above=-999 required=5 tests=[AWL=0.325, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id WmxijqEhjj58 for ; Fri, 14 Oct 2011 11:07:46 -0700 (PDT) Received: from mailer1.neclab.eu (mailer1.neclab.eu [195.37.70.40]) by ietfa.amsl.com (Postfix) with ESMTP id 8A64121F8CB2 for ; Fri, 14 Oct 2011 11:07:46 -0700 (PDT) Received: from localhost (localhost.localdomain [127.0.0.1]) by mailer1.neclab.eu (Postfix) with ESMTP id B4D56280001AC; Fri, 14 Oct 2011 20:07:45 +0200 (CEST) X-Virus-Scanned: Amavisd on Debian GNU/Linux (atlas1.office.hd) Received: from mailer1.neclab.eu ([127.0.0.1]) by localhost (atlas1.office.hd [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5HSt3Qv4R14I; Fri, 14 Oct 2011 20:07:45 +0200 (CEST) Received: from METHONE.office.hd (Methone.office.hd [192.168.24.54]) by mailer1.neclab.eu (Postfix) with ESMTP id 99FD6280001AB; Fri, 14 Oct 2011 20:07:35 +0200 (CEST) Received: from DAPHNIS.office.hd ([169.254.2.12]) by METHONE.office.hd ([192.168.24.54]) with mapi id 14.01.0323.003; Fri, 14 Oct 2011 20:07:35 +0200 From: Rolf Winter To: Padma Bhooma Thread-Topic: [ledbat] WG last call on draft-ietf-ledbat-congestion-08 Thread-Index: AQHMiG8gzp3UF1MPuUGreo3d3IkNOJV8J27M Date: Fri, 14 Oct 2011 18:07:35 +0000 Message-ID: <8D52A245-9713-4EB7-AE30-CA4B569FF774@office.hd> References: <2D481290-12BA-4101-81A1-07FF95A7CD29@apple.com> In-Reply-To: <2D481290-12BA-4101-81A1-07FF95A7CD29@apple.com> Accept-Language: en-US, de-DE Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "ledbat@ietf.org" Subject: Re: [ledbat] WG last call on draft-ietf-ledbat-congestion-08 X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 14 Oct 2011 18:07:47 -0000 This is awesome news. Have you done measurements with the implementation th= at you can share? On 12.10.2011, at 01:40, "Padma Bhooma" wrote: > Hi, >=20 > I reviewed the latest draft and I have few comments on the latest changes= . A receiver may have to echo the remote timestamp to allow for computation= of RTT which it can use for computing CTO. It might be useful to mention t= hat the sender is supposed to retransmit sent data or transmit new data to = restart the ACK clock. >=20 > Other than the comments above I think the algorithm is ready for adoption= . >=20 > The implementation of LEDBAT algorithm that was done as a TCP Congestion = control module in Mac OS X is available at the following link: > http://opensource.apple.com/source/xnu/xnu-1699.22.81/bsd/netinet/tcp_led= bat.c >=20 > Thanks, > Padma >=20 >=20 >=20 >=20 > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat From gorry@erg.abdn.ac.uk Sat Oct 15 00:40:54 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F0FDB21F8488 for ; Sat, 15 Oct 2011 00:40:54 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.299 X-Spam-Level: X-Spam-Status: No, score=-102.299 tagged_above=-999 required=5 tests=[AWL=-0.300, BAYES_00=-2.599, J_CHICKENPOX_43=0.6, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BDVFH7vrz5ft for ; Sat, 15 Oct 2011 00:40:54 -0700 (PDT) Received: from erg.abdn.ac.uk (dee.erg.abdn.ac.uk [IPv6:2001:630:241:204:203:baff:fe9a:8c9b]) by ietfa.amsl.com (Postfix) with ESMTP id 0602E21F8487 for ; Sat, 15 Oct 2011 00:40:53 -0700 (PDT) Received: from ra-gorry.erg.abdn.ac.uk (ra-gorry.erg.abdn.ac.uk [139.133.204.42]) by erg.abdn.ac.uk (8.13.4/8.13.4) with ESMTP id p9F7ef3j014877; Sat, 15 Oct 2011 08:40:42 +0100 (BST) Message-ID: <4E993917.5030508@erg.abdn.ac.uk> Date: Sat, 15 Oct 2011 08:41:11 +0100 From: Gorry Fairhurst Organization: University of Aberdeen User-Agent: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.7; en-US; rv:1.9.2.8) Gecko/20100802 Thunderbird/3.1.2 MIME-Version: 1.0 To: Joe Touch References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> <4E932F77.5060006@isi.edu> In-Reply-To: <4E932F77.5060006@isi.edu> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ERG-MailScanner: Found to be clean X-ERG-MailScanner-From: gorry@erg.abdn.ac.uk Cc: Janardhan Iyengar , ledbat@ietf.org Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: gorry@erg.abdn.ac.uk List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 15 Oct 2011 07:40:55 -0000 The latest text for the LEDBAT draft says: INIT_CWND SHOULD be 4, and MIN_CWND SHOULD be 2. An INIT_CWND of 4 should help seed FILTER() at the sender when there are no samples at the beginning of a flow, and a MIN_CWND of 2 allows FILTER() to use more than a single instantaneous delay estimate while not being too aggressive. Slight deviations may be warranted, for example, when these values of MIN_CWND and INIT_CWND interact poorly with the framing protocol. I commented previously on this, and here are my thoughts on the new text. I would personally be OK with starting with a window of 4. I do not agree that an update to TCP should be tracked by LEDBAT in the general case, because as I understand, the value 4 is derived from the operation of LEDBAT itself - i.e. use of the filter. In general, I like the idea of explaining when the SHOULD may be violated and what the implications are. However, the present clause at the end of this para seems to suggest ANY other value could also be OK, I do not agree this is to be left to the implementer - I was hoping here to see this limited to a fraction of the current TCP initial window, e.g. "but SHOULD NOT be larger than Max (4, one half of the initial window specified for TCP, currently [RFC3390])." However, I suspect that starting small is a very good idea for a longer-lived background flow that intends to be less aggressive than other flows, and should have little impact on the overall download time, therefore it may be better to just leave the recommendation as should use 4 in the first RTT. Gorry On 10/10/2011 18:46, Joe Touch wrote: > > > On 10/10/2011 9:46 AM, Mirja Kuehlewind wrote: >> Hi, >> >> I wouldn't want to refer to TCP's value because of a >> less-than-best-effort >> traffic there might always be a motivation to take different/lower >> values >> than TCP. > > The values currently suggested are TCP's current values. I don't mind > if you want to pick smaller ones, but I'd suggest spec'ing them > relative to TCP's current values regardless. > > Joe > >> >> Mirja >> >> >> On Monday 10 October 2011 16:37:31 Joe Touch wrote: >>> On 10/9/2011 4:51 PM, Janardhan Iyengar wrote: >>>> Dear all, >>>> >>>> A new version of the congestion control draft is in the repository. >>>> There are two major mods in this revision: >>> >>> ... >>> >>>> 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and >>>> have >>>> clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in >>>> Section 3.5. >>> >>> It might be more stable to refer to TCP's INIT_CWND and MIN_CWND >>> values, >>> i.e., that ledbat should use TCP's values for these. >>> >>> (they've been under discussion for revision in TCPM, FWIW) >>> >>> Joe >>> _______________________________________________ >>> ledbat mailing list >>> ledbat@ietf.org >>> https://www.ietf.org/mailman/listinfo/ledbat >> >> >> > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat > > From arjuna.sathiaseelan@gmail.com Sat Oct 15 01:37:08 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4823D21F8B1D for ; Sat, 15 Oct 2011 01:37:08 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.977 X-Spam-Level: X-Spam-Status: No, score=-2.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id VW2cOOAgwJkV for ; Sat, 15 Oct 2011 01:37:07 -0700 (PDT) Received: from mail-qw0-f44.google.com (mail-qw0-f44.google.com [209.85.216.44]) by ietfa.amsl.com (Postfix) with ESMTP id 2F6B321F8B1E for ; Sat, 15 Oct 2011 01:37:07 -0700 (PDT) Received: by qadb12 with SMTP id b12so1712873qad.31 for ; Sat, 15 Oct 2011 01:37:06 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=s0B8qJmItpR3mBysrk2kSjV1+JNbr8erJlkTilir+WA=; b=dcCozNym/40xa+p4EHLE8IejWFsBEOOosGxibXpNwMNXqEuC5SU807X9mlAj59Wy4j fSc7s7aRzYhNHMilDk8xk44FTkq5qxqv1zsPys+aVOPuemWjATh4f9bgiNoT2TG5xeqA Kfol3YbCki9WLJWPSdQ/BItQ/iuNLp9vqSkHw= MIME-Version: 1.0 Received: by 10.229.174.170 with SMTP id t42mr2519913qcz.137.1318667826633; Sat, 15 Oct 2011 01:37:06 -0700 (PDT) Sender: arjuna.sathiaseelan@gmail.com Received: by 10.229.224.69 with HTTP; Sat, 15 Oct 2011 01:37:06 -0700 (PDT) In-Reply-To: References: <4E92338F.1030601@fandm.edu> Date: Sat, 15 Oct 2011 09:37:06 +0100 X-Google-Sender-Auth: labm-nq6L3JDERgPMknbpnVEEiU Message-ID: From: Arjuna Sathiaseelan To: janardhan.iyengar@fandm.edu Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: adithya kumar , ledbat@ietf.org, "Dr. R.Leela Velusamy" Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 15 Oct 2011 08:37:08 -0000 Dear Jana and others, =A0I went through the draft and am currently trying to figure out what the performance implications are with the latest draft through simulations (I have a few helpful UG students in India who have volunteered to do this for me - this includes modifying the current ns-2 code to reflect the latest draft). But unfortunately, I dont think we will be able to tell anything useful before WGLC ends! However I have one question -> =A0" A CTO is used to detect heavy congestion indicated by loss of all =A0 outstanding data or acknowledgments, resulting in reduction of the =A0 cwnd to 1 MSS and an exponential backoff of the CTO interval" So I am trying to see =A0whether heavy congestion actually means loss of "ALL" outstanding segments - so how does LEDBAT behave according to the current draft if atleast one packet went through? Just trying to understand. Regards Arjuna On 10 October 2011 00:51, Janardhan Iyengar wrote: > Dear all, > > A new version of the congestion control draft is in the repository. There > are two major mods in this revision: > > 1/ WGLC identified one major issue that needed to be addressed in the LED= BAT > congestion control draft -- LEDBAT response to extreme congestion -- and > we've tried to address that issue in this revision. We've added a new > mechanism, the Congestion Timeout (CTO), for a sender to respond to extre= me > congestion. We do not specify how this should be implemented, but we do n= ote > that a CTO can be implemented with or without a timer. In terms of textua= l > changes, we've added an update_CTO() function and a branch for what to do= if > no acks are received within a CTO amount of time in Section 3.4.2. We hav= e > changed the response to data loss to ensure that a protocol, such as TCP, > that uses the same timer for both congestion control and for > retransmissions, changes its cwnd correctly. > > 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and have > clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in Sect= ion > 3.5. > > Please comment! > - jana > > -- > Janardhan Iyengar > Assistant Professor, Computer Science > Franklin & Marshall College > http://www.fandm.edu/jiyengar > > > -------- Original Message -------- > Subject: New Version Notification for draft-ietf-ledbat-congestion-08.txt > Date: Sun, 09 Oct 2011 16:33:33 -0700 > From: internet-drafts@ietf.org > To: jiyengar@fandm.edu > CC: jiyengar@fandm.edu, mirja.kuehlewind@ikr.uni-stuttgart.de, > =A0greg@bittorrent.com, shalunov@bittorrent.com > > A new version of I-D, draft-ietf-ledbat-congestion-08.txt has been > successfully submitted by Janardhan Iyengar and posted to the IETF > repository. > > Filename: =A0 =A0 =A0 =A0draft-ietf-ledbat-congestion > Revision: =A0 =A0 =A0 =A008 > Title: =A0 =A0 =A0 =A0 =A0 Low Extra Delay Background Transport (LEDBAT) > Creation date: =A0 2011-10-09 > WG ID: =A0 =A0 =A0 =A0 =A0 ledbat > Number of pages: 19 > > Abstract: > =A0 LEDBAT is an experimental delay-based congestion control algorithm > =A0 that attempts to utilize the available bandwidth on an end-to-end > =A0 path while limiting the consequent increase in queueing delay on the > =A0 path. =A0LEDBAT uses changes in one-way delay measurements to limit > =A0 congestion that the flow itself induces in the network. =A0LEDBAT is > =A0 designed for use by background bulk-transfer applications; it is > =A0 designed to be no more aggressive than TCP congestion control and to > =A0 yield in the presence of any competing flows when latency builds, > =A0 thus limiting interference with the network performance of the > =A0 competing flows. > > > > > The IETF Secretariat > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat > -- http://about.me/arjuna.sathiaseelan From arjuna.sathiaseelan@gmail.com Sat Oct 15 01:56:50 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id E425821F8A7D for ; Sat, 15 Oct 2011 01:56:50 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.977 X-Spam-Level: X-Spam-Status: No, score=-2.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jT0sYcuF8RhO for ; Sat, 15 Oct 2011 01:56:50 -0700 (PDT) Received: from mail-qw0-f44.google.com (mail-qw0-f44.google.com [209.85.216.44]) by ietfa.amsl.com (Postfix) with ESMTP id D9D3B21F8A6C for ; Sat, 15 Oct 2011 01:56:49 -0700 (PDT) Received: by qadb12 with SMTP id b12so1717411qad.31 for ; Sat, 15 Oct 2011 01:56:49 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=Ndnx/mxjstPY03QjT2feGm8MJ6JUzoGHA2ktprzcQtQ=; b=ojzY3pjNig0Y6pocACaXh2ETNiMzIe/COl0Amn0rGbg4PSiHAhsddKilKQQ8N7Sacf VX9o9B1YJohmAOnN7rC2CCrinconSxF4eaHwFMlideS9Mhzlcu+5CuVeHhrJ2pjmcVKe jZHzEuRLjSL/cb6XxmRMJdDoSR8qZAsHL11cI= MIME-Version: 1.0 Received: by 10.229.51.137 with SMTP id d9mr2544609qcg.61.1318669009312; Sat, 15 Oct 2011 01:56:49 -0700 (PDT) Sender: arjuna.sathiaseelan@gmail.com Received: by 10.229.224.69 with HTTP; Sat, 15 Oct 2011 01:56:49 -0700 (PDT) In-Reply-To: References: <4E92338F.1030601@fandm.edu> Date: Sat, 15 Oct 2011 09:56:49 +0100 X-Google-Sender-Auth: JtBEEVMiOYa2EKAWtK8iJNHWSIQ Message-ID: From: Arjuna Sathiaseelan To: janardhan.iyengar@fandm.edu Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: ledbat@ietf.org Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 15 Oct 2011 08:56:51 -0000 Another thing which I forgot to point out - in section 3.4.1 the term off_target suddenly pops up without an explanation. That needs to be resolved please. Regards Arjuna On 15 October 2011 09:37, Arjuna Sathiaseelan wrote= : > Dear Jana and others, > > =A0I went through the draft and am currently trying to figure out what > the performance implications are with the latest draft through > simulations (I have a few helpful UG students in India who have > volunteered to do this for me - this includes modifying the current > ns-2 code to reflect the latest draft). But unfortunately, I dont > think we will be able to tell anything useful before WGLC ends! > > However I have one question -> > =A0" A CTO is used to detect heavy congestion indicated by loss of all > =A0 outstanding data or acknowledgments, resulting in reduction of the > =A0 cwnd to 1 MSS and an exponential backoff of the CTO interval" > > So I am trying to see =A0whether heavy congestion actually means loss of > "ALL" outstanding segments - so how does LEDBAT behave according to > the current draft if atleast one packet went through? Just trying to > understand. > > Regards > Arjuna > > On 10 October 2011 00:51, Janardhan Iyengar wrot= e: >> Dear all, >> >> A new version of the congestion control draft is in the repository. Ther= e >> are two major mods in this revision: >> >> 1/ WGLC identified one major issue that needed to be addressed in the LE= DBAT >> congestion control draft -- LEDBAT response to extreme congestion -- and >> we've tried to address that issue in this revision. We've added a new >> mechanism, the Congestion Timeout (CTO), for a sender to respond to extr= eme >> congestion. We do not specify how this should be implemented, but we do = note >> that a CTO can be implemented with or without a timer. In terms of textu= al >> changes, we've added an update_CTO() function and a branch for what to d= o if >> no acks are received within a CTO amount of time in Section 3.4.2. We ha= ve >> changed the response to data loss to ensure that a protocol, such as TCP= , >> that uses the same timer for both congestion control and for >> retransmissions, changes its cwnd correctly. >> >> 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and have >> clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in Sec= tion >> 3.5. >> >> Please comment! >> - jana >> >> -- >> Janardhan Iyengar >> Assistant Professor, Computer Science >> Franklin & Marshall College >> http://www.fandm.edu/jiyengar >> >> >> -------- Original Message -------- >> Subject: New Version Notification for draft-ietf-ledbat-congestion-08.tx= t >> Date: Sun, 09 Oct 2011 16:33:33 -0700 >> From: internet-drafts@ietf.org >> To: jiyengar@fandm.edu >> CC: jiyengar@fandm.edu, mirja.kuehlewind@ikr.uni-stuttgart.de, >> =A0greg@bittorrent.com, shalunov@bittorrent.com >> >> A new version of I-D, draft-ietf-ledbat-congestion-08.txt has been >> successfully submitted by Janardhan Iyengar and posted to the IETF >> repository. >> >> Filename: =A0 =A0 =A0 =A0draft-ietf-ledbat-congestion >> Revision: =A0 =A0 =A0 =A008 >> Title: =A0 =A0 =A0 =A0 =A0 Low Extra Delay Background Transport (LEDBAT) >> Creation date: =A0 2011-10-09 >> WG ID: =A0 =A0 =A0 =A0 =A0 ledbat >> Number of pages: 19 >> >> Abstract: >> =A0 LEDBAT is an experimental delay-based congestion control algorithm >> =A0 that attempts to utilize the available bandwidth on an end-to-end >> =A0 path while limiting the consequent increase in queueing delay on the >> =A0 path. =A0LEDBAT uses changes in one-way delay measurements to limit >> =A0 congestion that the flow itself induces in the network. =A0LEDBAT is >> =A0 designed for use by background bulk-transfer applications; it is >> =A0 designed to be no more aggressive than TCP congestion control and to >> =A0 yield in the presence of any competing flows when latency builds, >> =A0 thus limiting interference with the network performance of the >> =A0 competing flows. >> >> >> >> >> The IETF Secretariat >> _______________________________________________ >> ledbat mailing list >> ledbat@ietf.org >> https://www.ietf.org/mailman/listinfo/ledbat >> > > > > -- > http://about.me/arjuna.sathiaseelan > From touch@isi.edu Mon Oct 17 07:10:39 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CF9E121F8BA8 for ; Mon, 17 Oct 2011 07:10:39 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -104.299 X-Spam-Level: X-Spam-Status: No, score=-104.299 tagged_above=-999 required=5 tests=[AWL=1.700, BAYES_00=-2.599, J_CHICKENPOX_43=0.6, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id R12AJMneWiDI for ; Mon, 17 Oct 2011 07:10:39 -0700 (PDT) Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by ietfa.amsl.com (Postfix) with ESMTP id E66EC21F8B58 for ; Mon, 17 Oct 2011 07:10:38 -0700 (PDT) Received: from [192.168.1.94] (pool-71-105-89-105.lsanca.dsl-w.verizon.net [71.105.89.105]) (authenticated bits=0) by boreas.isi.edu (8.13.8/8.13.8) with ESMTP id p9HE98OR028227 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Mon, 17 Oct 2011 07:09:19 -0700 (PDT) Mime-Version: 1.0 (Apple Message framework v1244.3) Content-Type: text/plain; charset=iso-8859-1 From: Joe Touch In-Reply-To: <4E993917.5030508@erg.abdn.ac.uk> Date: Mon, 17 Oct 2011 07:09:08 -0700 Content-Transfer-Encoding: quoted-printable Message-Id: References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> <4E932F77.5060006@isi.edu> <4E993917.5030508@erg.abdn.ac.uk> To: gorry@erg.abdn.ac.uk X-Mailer: Apple Mail (2.1244.3) X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: Janardhan Iyengar , ledbat@ietf.org Subject: Re: [ledbat] New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 17 Oct 2011 14:10:40 -0000 Hi, Gorry (et al.), On Oct 15, 2011, at 12:41 AM, Gorry Fairhurst wrote: > The latest text for the LEDBAT draft says: >=20 > INIT_CWND SHOULD be 4, and MIN_CWND SHOULD be 2. An INIT_CWND of 4 > should help seed FILTER() at the sender when there are no samples at > the beginning of a flow, and a MIN_CWND of 2 allows FILTER() to use > more than a single instantaneous delay estimate while not being too > aggressive. Slight deviations may be warranted, for example, when > these values of MIN_CWND and INIT_CWND interact poorly with the > framing protocol. >=20 > I commented previously on this, and here are my thoughts on the new = text. >=20 > I would personally be OK with starting with a window of 4. >=20 > I do not agree that an update to TCP should be tracked by LEDBAT in = the general case, because as I understand, the value 4 is derived from = the operation of LEDBAT itself - i.e. use of the filter. Sec 3.2: In addition to the LEDBAT mechanism described below, we note that a slow start mechanism can be used as specified in [RFC5681]. Since slow start leads to faster increase in the window than that specified in LEDBAT, conservative congestion control implementations employing LEDBAT may skip slow start altogether and start with an initial window of INIT_CWND * MSS. (INIT_CWND is described later in Section 3.5.) Sec 3.5: A LEDBAT sender uses the current_delays list to maintain only delay measurements made within a RTT amount of time in the past, seeking to eliminate noise spikes in its measurement of the current one-way delay through the network. The size of this list, CURRENT_FILTER, may be variable, and depends on the number of successful measurements made within a RTT amount of time in the past. The sender should seek to gather enough delay samples in each RTT so as to have statistical confidence in the measurements. While the number of delay samples required for such confidence will vary depending on network conditions, we recommend that the sender SHOULD use at least 4 samples in each RTT, unless the number of samples is lower due to a small congestion window. Thus, subject to congestion window constraints, the number of delay samples in each RTT SHOULD be at least 4, and thus CURRENT_FILTER SHOULD be at least 4. CURRENT_FILTER MUST be limited such that samples in the list are not older than an RTT in the past. INIT_CWND SHOULD be 4, and MIN_CWND SHOULD be 2. An INIT_CWND of 4 should help seed FILTER() at the sender when there are no samples at the beginning of a flow, and a MIN_CWND of 2 allows FILTER() to use more than a single instantaneous delay estimate while not being too aggressive. Slight deviations may be warranted, for example, when these values of MIN_CWND and INIT_CWND interact poorly with the framing protocol. For INIT_CWND, the number 4 is derived from a notion of the minimum = number of samples needed to make FILTER work. However, it clearly should = not be larger than TCP's INIT_CWND, and the number 4 appears to be = related to a misunderstanding of TCP's initial value (see below), not = picked solely for its ability to generate meaningful statistics. IMO, the right description would be: no smaller than needed to ensure a reasonable number of samples no larger than TCP's current values Note that the definition of "reasonable number of samples" can be = described more clearly as, e.g., SHOULD use enough saamples to yield a variance of less than 25%=20= I'm not sure whether "variance" or some other confidence interval should = be specified, or whether 25% is better than other values. However, the = current text is vague on both points. > In general, I like the idea of explaining when the SHOULD may be = violated and what the implications are. However, the present clause at = the end of this para seems to suggest ANY other value could also be OK, = I do not agree this is to be left to the implementer - I was hoping here = to see this limited to a fraction of the current TCP initial window, = e.g. "but SHOULD NOT be larger than Max (4, one half of the initial = window specified for TCP, currently [RFC3390])." However, I suspect that = starting small is a very good idea for a longer-lived background flow = that intends to be less aggressive than other flows, and should have = little impact on the overall download time, therefore it may be better = to just leave the recommendation as should use 4 in the first RTT. >=20 > Gorry Unfortunately, the equation above yields a current value of 1 (RFC3990 = is 4 MSS *or* 4KB, whichever is smaller, and for 1500-byte MTUs that is = 3; half of 3 would round down to 1). Even picking TCP's current INIT_CWND should be 3, not 4, based on = typical MSS sizes. So this analysis indicates a big problem with INIT_CWND in LEDBAT - the = smallest viable value needs to be large enough to have a meaningful = average, but lower than TCP's INIT_CWND, and there doesn't appear to be = a currently viable value. Further, INIT_CWND should always be divisible by the ACK frequency = (e.g., if 1 ACK per 2 MSS, then divisible by 2), to avoid unnecessary = delays due to the ACK frequency compression. MIN_CWND IMO should be defined as half INIT_CWND, *NOT* as a fixed value = in this doc. Joe > On 10/10/2011 18:46, Joe Touch wrote: >>=20 >>=20 >> On 10/10/2011 9:46 AM, Mirja Kuehlewind wrote: >>> Hi, >>>=20 >>> I wouldn't want to refer to TCP's value because of a = less-than-best-effort >>> traffic there might always be a motivation to take different/lower = values >>> than TCP. >>=20 >> The values currently suggested are TCP's current values. I don't mind = if you want to pick smaller ones, but I'd suggest spec'ing them relative = to TCP's current values regardless. >>=20 >> Joe >>=20 >>>=20 >>> Mirja >>>=20 >>>=20 >>> On Monday 10 October 2011 16:37:31 Joe Touch wrote: >>>> On 10/9/2011 4:51 PM, Janardhan Iyengar wrote: >>>>> Dear all, >>>>>=20 >>>>> A new version of the congestion control draft is in the = repository. >>>>> There are two major mods in this revision: >>>>=20 >>>> ... >>>>=20 >>>>> 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, = and have >>>>> clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND = in >>>>> Section 3.5. >>>>=20 >>>> It might be more stable to refer to TCP's INIT_CWND and MIN_CWND = values, >>>> i.e., that ledbat should use TCP's values for these. >>>>=20 >>>> (they've been under discussion for revision in TCPM, FWIW) >>>>=20 >>>> Joe >>>> _______________________________________________ >>>> ledbat mailing list >>>> ledbat@ietf.org >>>> https://www.ietf.org/mailman/listinfo/ledbat >>>=20 >>>=20 >>>=20 >> _______________________________________________ >> ledbat mailing list >> ledbat@ietf.org >> https://www.ietf.org/mailman/listinfo/ledbat >>=20 >>=20 From David.Ros@telecom-bretagne.eu Tue Oct 18 08:30:05 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3215721F8BEF for ; Tue, 18 Oct 2011 08:30:05 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.649 X-Spam-Level: X-Spam-Status: No, score=-1.649 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HELO_EQ_FR=0.35, J_CHICKENPOX_43=0.6] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id ZN5TmszBqCVL for ; Tue, 18 Oct 2011 08:30:04 -0700 (PDT) Received: from coliposte.enst-bretagne.fr (coliposte.enst-bretagne.fr [192.108.115.12]) by ietfa.amsl.com (Postfix) with ESMTP id 35B2621F8BE9 for ; Tue, 18 Oct 2011 08:30:04 -0700 (PDT) Received: from localhost (localhost.localdomain [127.0.0.1]) by coliposte.enst-bretagne.fr (8.13.7/8.13.7/2009.11.10) with ESMTP id p9IFTwVv016054; Tue, 18 Oct 2011 17:29:59 +0200 Received: from courrier.enst-bretagne.fr (smtps.telecom-bretagne.eu [10.29.90.4]) by coliposte.enst-bretagne.fr (8.13.7/8.13.7/2009.11.10) with ESMTP id p9IFTs9O016040; Tue, 18 Oct 2011 17:29:57 +0200 Received: from eduroam-193-157-110-184.uio.no (passerelle-interne.enst-bretagne.fr [192.108.117.210]) (user=dros mech=PLAIN bits=0) by courrier.enst-bretagne.fr (8.13.8/8.13.8/2010.02.22) with ESMTP id p9IFTnBb009705; Tue, 18 Oct 2011 17:29:49 +0200 Mime-Version: 1.0 (Apple Message framework v1084) Content-Type: text/plain; charset=iso-8859-1 From: David Ros In-Reply-To: Date: Tue, 18 Oct 2011 17:29:49 +0200 Content-Transfer-Encoding: quoted-printable Message-Id: References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> <4E932F77.5060006@isi.edu> <4E993917.5030508@erg.abdn.ac.uk> To: Joe Touch X-Mailer: Apple Mail (2.1084) X-Virus-Scanned: amavisd-new at enst-bretagne.fr Cc: Janardhan Iyengar , ledbat@ietf.org Subject: Re: [ledbat] New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Oct 2011 15:30:05 -0000 Le 17 oct. 2011 =E0 16:09, Joe Touch a =E9crit : > [...] > Unfortunately, the equation above yields a current value of 1 (RFC3990 = is 4 MSS *or* 4KB, whichever is smaller, and for 1500-byte MTUs that is = 3; half of 3 would round down to 1). >=20 > Even picking TCP's current INIT_CWND should be 3, not 4, based on = typical MSS sizes. >=20 > So this analysis indicates a big problem with INIT_CWND in LEDBAT - = the smallest viable value needs to be large enough to have a meaningful = average, but lower than TCP's INIT_CWND, and there doesn't appear to be = a currently viable value. Hi all, Sorry if I'm getting this totally wrong, but: is it really *necessary* = that LEDBAT's INIT_CWND be *smaller* than TCP's? As far as I understand, = the point of LEDBAT is to be less-than-best-effort over (relatively) = long timescales, or at least over time scales longer than one (initial) = RTT. And just starting up as TCP won't make it *more* aggressive than = TCP. Is this correct?? (of course, this doesn't change anything to other points made by Joe -- = e.g., if LEDBAT uses INIT_CWND =3D 4 then it'll start by sending more = than what TCP would -- i.e., 3 segments -- with typical MSS values) Thanks, David. >=20 > Further, INIT_CWND should always be divisible by the ACK frequency = (e.g., if 1 ACK per 2 MSS, then divisible by 2), to avoid unnecessary = delays due to the ACK frequency compression. >=20 > MIN_CWND IMO should be defined as half INIT_CWND, *NOT* as a fixed = value in this doc. >=20 > Joe >=20 >=20 >=20 >> On 10/10/2011 18:46, Joe Touch wrote: >>>=20 >>>=20 >>> On 10/10/2011 9:46 AM, Mirja Kuehlewind wrote: >>>> Hi, >>>>=20 >>>> I wouldn't want to refer to TCP's value because of a = less-than-best-effort >>>> traffic there might always be a motivation to take different/lower = values >>>> than TCP. >>>=20 >>> The values currently suggested are TCP's current values. I don't = mind if you want to pick smaller ones, but I'd suggest spec'ing them = relative to TCP's current values regardless. >>>=20 >>> Joe >>>=20 >>>>=20 >>>> Mirja >>>>=20 >>>>=20 >>>> On Monday 10 October 2011 16:37:31 Joe Touch wrote: >>>>> On 10/9/2011 4:51 PM, Janardhan Iyengar wrote: >>>>>> Dear all, >>>>>>=20 >>>>>> A new version of the congestion control draft is in the = repository. >>>>>> There are two major mods in this revision: >>>>>=20 >>>>> ... >>>>>=20 >>>>>> 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, = and have >>>>>> clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND = in >>>>>> Section 3.5. >>>>>=20 >>>>> It might be more stable to refer to TCP's INIT_CWND and MIN_CWND = values, >>>>> i.e., that ledbat should use TCP's values for these. >>>>>=20 >>>>> (they've been under discussion for revision in TCPM, FWIW) >>>>>=20 >>>>> Joe >>>>> _______________________________________________ >>>>> ledbat mailing list >>>>> ledbat@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/ledbat >>>>=20 >>>>=20 >>>>=20 >>> _______________________________________________ >>> ledbat mailing list >>> ledbat@ietf.org >>> https://www.ietf.org/mailman/listinfo/ledbat >>>=20 >>>=20 >=20 > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat >=20 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D David ROS http://www.rennes.enst-bretagne.fr/~dros/ In theory, theory and practice are the same. In practice, they are = different. -- Author unknown From touch@isi.edu Tue Oct 18 12:28:09 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2F8DA21F8CEA for ; Tue, 18 Oct 2011 12:28:09 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.599 X-Spam-Level: X-Spam-Status: No, score=-106.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UVSI+Tp0JMkh for ; Tue, 18 Oct 2011 12:28:08 -0700 (PDT) Received: from boreas.isi.edu (boreas.isi.edu [128.9.160.161]) by ietfa.amsl.com (Postfix) with ESMTP id B034521F8CDF for ; Tue, 18 Oct 2011 12:28:08 -0700 (PDT) Received: from [172.16.38.64] (205.158.58.226.ptr.us.xo.net [205.158.58.226]) (authenticated bits=0) by boreas.isi.edu (8.13.8/8.13.8) with ESMTP id p9IJR1WJ004756 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Tue, 18 Oct 2011 12:27:11 -0700 (PDT) Message-ID: <4E9DD306.4020402@isi.edu> Date: Tue, 18 Oct 2011 12:27:02 -0700 From: Joe Touch User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1 MIME-Version: 1.0 To: David Ros References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> <4E932F77.5060006@isi.edu> <4E993917.5030508@erg.abdn.ac.uk> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: Janardhan Iyengar , ledbat@ietf.org Subject: Re: [ledbat] New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Oct 2011 19:28:09 -0000 On 10/18/2011 8:29 AM, David Ros wrote: ... > Hi all, > > Sorry if I'm getting this totally wrong, but: is it really > *necessary* > that LEDBAT's INIT_CWND be *smaller* than TCP's? As far as I understand, > the point of LEDBAT is to be less-than-best-effort over (relatively) > long timescales, or at least over time scales longer than one (initial) > RTT. And just starting up as TCP won't make it *more* aggressive than > TCP. Is this correct?? No, but it could make it a lot like TCP if the offered load is in short bursts. If that's not the intent, then the INIT_CWND needs to be smaller than TCP's. Joe From bhooma@apple.com Tue Oct 18 22:00:24 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B753F11E8083 for ; Tue, 18 Oct 2011 22:00:24 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.426 X-Spam-Level: X-Spam-Status: No, score=-106.426 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_IMAGE_RATIO_04=0.172, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UwVRXAWHfetA for ; Tue, 18 Oct 2011 22:00:23 -0700 (PDT) Received: from mail-out.apple.com (mail-out.apple.com [17.151.62.49]) by ietfa.amsl.com (Postfix) with ESMTP id 2261511E8080 for ; Tue, 18 Oct 2011 22:00:19 -0700 (PDT) MIME-version: 1.0 Content-type: multipart/alternative; boundary="Boundary_(ID_+ckLzYkDTicPnCgDPFpFXw)" Received: from relay11.apple.com ([17.128.113.48]) by mail-out.apple.com (Oracle Communications Messaging Exchange Server 7u4-20.01 64bit (built Nov 21 2010)) with ESMTPS id <0LTA00K0UR8B5BM1@mail-out.apple.com> for ledbat@ietf.org; Tue, 18 Oct 2011 22:00:17 -0700 (PDT) X-AuditID: 11807130-b7c1bae000001553-37-4e9e595fe588 Received: from [17.153.35.95] (Unknown_Domain [17.153.35.95]) by relay11.apple.com (Apple SCV relay) with SMTP id 0A.A2.05459.0695E9E4; Tue, 18 Oct 2011 22:00:17 -0700 (PDT) From: Padma Bhooma In-reply-to: <8D52A245-9713-4EB7-AE30-CA4B569FF774@office.hd> Date: Tue, 18 Oct 2011 22:00:16 -0700 Message-id: References: <2D481290-12BA-4101-81A1-07FF95A7CD29@apple.com> <8D52A245-9713-4EB7-AE30-CA4B569FF774@office.hd> To: Rolf Winter X-Mailer: Apple Mail (2.1247) X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFrrMIsWRmVeSWpSXmKPExsUiOFM5Xjcxcp6fwZo5Ehb/Hj1ns3h/9yWr A5PHkiU/mTxmHHvJHsAUxWWTkpqTWZZapG+XwJVxrEmo4Okx9orfjzeyNzB+WMbexcjBISFg IjHth0gXIyeQKSZx4d56ti5GLg4hgfWMEj9nv2AGqREWcJWY06kMUsMrYCjxc/I6JhCbWSBB YtH1lewgNpuAqsTJP3eYQWxOAVuJhqMvwGwWoPizSU+YIerVJd73zmCHmGMjceHdXzYQW0gg V2L91SesILYIUM2S96dYIe6RlXj39g3bBEa+WUhWz0KyGsLWlli28DXzLKBLmQV0JCYvZIQI F0n8uP8ATQmE/fH8EagxihJTuh9C1ZhLfPyznRXCTpT4eqSNBWKki8SRb+aoykHCphJHLkpA hOMlVnd1s8JUdx8KxDTcTOL/xW1QB7hI/Luwgg3GXrx6B9sCRuFVjIJFqTmJlYaGeokFBTmp esn5uZsYQVHdUGiwg3HtT/5DjAIcjEo8vFxS8/yEWBPLiitzDzFKcDArifCujwAK8aYkVlal FuXHF5XmpBYfYpTmYFES5+XvmesnJJCeWJKanZpakFoEk2Xi4JRqYDQ4oRn6dKVAYfC0qd/L 7jMZTp+jPvPc9Lpbap7e/Q/n7bYUk1tT/FCh5Fh7yArDwNWFkhOWWxrsXPP+aOrVYtVJU6ek L2pWSfAMTc9t557KmrFhQeeyX+LrfhxbdXw2q8Bz7Vm5DOW3akIm6i/OP5RulN1g4+uyKUHk WN1DTkUH4f6OOI9IDSWW4oxEQy3mouJEAG0Y+eDmAgAA Cc: "ledbat@ietf.org" Subject: Re: [ledbat] WG last call on draft-ietf-ledbat-congestion-08 X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 19 Oct 2011 05:00:24 -0000 --Boundary_(ID_+ckLzYkDTicPnCgDPFpFXw) Content-type: text/plain; CHARSET=US-ASCII Content-transfer-encoding: 7BIT We have done measurements to see how LEDBAT will behave on wireless links under different kinds of network conditions. Our main emphasis was to see improvement in response time and throughput for a competing flow when there is a LEDBAT upload in progress compared to a standard TCP flow in progress. For instance, the following graph shows how web page load times to different websites changed when a standard TCP upload (blue) or LEDBAT (green) or no upload flow (yellow) was in progress to another host. The numbers with LEDBAT are very close to having no upload and they show a huge improvement in response time. The Y-axis is page load times in milliseconds. Here is another graph for similar measurements on 3G: Another important goal for LEDBAT is to make sure that a LEDBAT flow will yield to a standard TCP flow when they have to share the link. Here is a graph to show how two LEDBAT flows (Flow A and Flow B marked as bk) yielded to another loss-based TCP flow: Overall, we have observed that using LEDBAT as a congestion control algorithm improves response time for other applications which is very desirable in some scenarios. Thanks, Padma On Oct 14, 2011, at 11:07 AM, Rolf Winter wrote: > This is awesome news. Have you done measurements with the implementation that you can share? > > > > On 12.10.2011, at 01:40, "Padma Bhooma" wrote: > >> Hi, >> >> I reviewed the latest draft and I have few comments on the latest changes. A receiver may have to echo the remote timestamp to allow for computation of RTT which it can use for computing CTO. It might be useful to mention that the sender is supposed to retransmit sent data or transmit new data to restart the ACK clock. >> >> Other than the comments above I think the algorithm is ready for adoption. >> >> The implementation of LEDBAT algorithm that was done as a TCP Congestion control module in Mac OS X is available at the following link: >> http://opensource.apple.com/source/xnu/xnu-1699.22.81/bsd/netinet/tcp_ledbat.c >> >> Thanks, >> Padma >> >> >> >> >> _______________________________________________ >> ledbat mailing list >> ledbat@ietf.org >> https://www.ietf.org/mailman/listinfo/ledbat --Boundary_(ID_+ckLzYkDTicPnCgDPFpFXw) Content-type: multipart/related; boundary="Boundary_(ID_UQnfOsK19lMNJYbtgzxRtQ)"; type="text/html" --Boundary_(ID_UQnfOsK19lMNJYbtgzxRtQ) Content-type: text/html; CHARSET=US-ASCII Content-transfer-encoding: quoted-printable We = have done measurements to see how LEDBAT will behave on wireless links = under different kinds of network conditions. Our main
emphasis was = to see improvement in response time and throughput for a competing flow = when there is a LEDBAT upload in 
progress compared to a = standard TCP flow in progress. 

For = instance, the following graph shows how web page load times to different = websites changed when a standard TCP upload (blue)
or LEDBAT = (green) or no upload flow (yellow) was in progress to another host. The = numbers with LEDBAT are very close to having
no upload and = they show a huge improvement in response time. The Y-axis is page load = times in = milliseconds.




<= /div>
<= div>
Here is another graph for similar measurements on = 3G:


<= div>
Another important goal for LEDBAT is to make sure = that a LEDBAT flow will yield to a standard TCP flow when they have to = share 
the link. Here is a graph to show how two LEDBAT = flows (Flow A and Flow B marked as bk) yielded to another loss-based TCP = flow:


<= div>
Overall, we have observed that using LEDBAT as a = congestion control algorithm improves response time for = other 
applications which is very desirable in some = scenarios.

Thanks,
Padma

=

On Oct 14, 2011, at 11:07 AM, = Rolf Winter wrote:

This = is awesome news. Have you done measurements with the implementation that = you can share?



On 12.10.2011, at 01:40, "Padma Bhooma" = <bhooma@apple.com> = wrote:

Hi,

I reviewed the = latest draft and I have few comments on the latest changes. A receiver = may have to echo the remote timestamp to allow for computation of RTT = which it can use for computing CTO. It might be useful to mention that = the sender is supposed to retransmit sent data or transmit new data to = restart the ACK clock.

Other than the = comments above I think the algorithm is ready for = adoption.

The = implementation of LEDBAT algorithm that was done as a TCP Congestion = control module in Mac OS X is available at the following = link:
http://opensource.apple.com/source/xnu/xnu-1699.22.81/bsd/ne= tinet/tcp_ledbat.c

Thanks,
Padma




_______________________________________________
ledbat mailing = list
ledbat@ietf.org
https://www.ietf.org= /mailman/listinfo/ledbat

=
= --Boundary_(ID_UQnfOsK19lMNJYbtgzxRtQ) Content-id: Content-type: image/png; x-mac-hide-extension=yes; x-unix-mode=0644; name=wifi-upload.png Content-transfer-encoding: base64 Content-disposition: inline; filename=wifi-upload.png iVBORw0KGgoAAAANSUhEUgAAAekAAAGSCAIAAABISa7CAAAXfmlDQ1BJQ0MgUHJvZmlsZQAAeAGt WWVYVU3Xnr1Pcjh0d3d3SXd3CsLh0M0hRCxEREJBUBABURBBpASUMigBEQMMbEJAKREVBAHl26A+ 7/Pjff99+7rO3jdr7rVmzazZa88aAOC4ToqKCoPpAQiPiKU4mBrwu7l78OPeABxgAYxAEvCRyDFR +nZ2VuB/Xt9HALTd+FRm29b/pP33BgY//xgyAJAd0uzrF0MOR/B1AOBmchQlFgD0tj2hvbFR2/gU gpkpiIMILt/Ggb9x6zb2/Y0f7HCcHAwRzjgAeCKJRAkEgLiAyPnjyYGIHRoiAFjGCL/gCACY+BGs Qw4i+QHA4YNwpMPDI7fxCQSL+/7LTuC/MInk+49NEinwH/x7LIgm0rFRcExUGGnfzh//n7fwsDhk vnYuVuROjIo1cECe7Mi8sQfHmjshmBnB8kFxZs5/sFFikJPrNheRu0X42tgimBHBfuQYQ2QuAWIH ig+NtNy2s8054edvZIxgZFVABTHxjn/xlcQgQ5s/HLcQksV2zGgRTiuJgqDf/fZFxdpt+7Bt83VE mI3VH7wUQDHZto/IYax/jLEjghEfYO5YitO2HPEZlg0INjFHMNIvbBAVtrPmtjkOlDiH7bEII9jP P8L5r26aH8nIEpFzI/JCYAUMgRHgR+6RIAz5UUAw8EOef+Xkf8kdQSL4BCKAP4hBNHYY3sHJlL8Y mAASoh+ItMv80TfYkfiDeETr51/eo4WWhb/4j47vPxomYGrHxh8L8tXys/Kbf9n8dH/9whpjjbBm WBOsxF8J0tPvUVB2/LNERuMP4hBb/kjff/3596ji/mH8W/p7Dhx2tEIRRvDfvoHLjmfB/9iy/Gdm /swFWhStiFZBG6C10TpoDcCPZkVzAhm0MlodrY/WRWshbRr/muc/Wn/8lwEBO3MVv+N9KPiAeI68 1bH+CbFIrIBhZNQ+SnBgUCy/PpIt/KX5zSPIstL8ivIKCmA792xzAPjmsJNTINbH/5H5vwFA0wQA qvL/yIIGAaipBoCz9D8ycXnk/aUDoG6GHEeJ/20Pvf3AAAKgQ1YaB+AFQkAcGb8iUAVaQA8YAwtg C5yAO9gDyCAI8ZcC9oID4AhIBZngFMgDheACuASugFrQCFrALdAF+sAgGALPwRswDqbBPFgC38EG BEE4iAZigjggPkgEkoIUIXVIBzKGrCAHyB3ygQKhCCgOOgAdhTKhXKgQKoWqoAaoDeqCBqBh6BU0 Ac1CX6F1GAUTYWaYBxaF5WB1WB+2hJ1gLzgQjoYT4RQ4Cy6Ay+AauBnuggfh5/A4PA+voACKGsWK EkDJoNRRhihblAcqAEVBHUJloPJRZag6VDuqH/UUNY5aQP1AY9FMaH60DBJLM7QzmoyORh9Cn0AX oq+gm9F30U/RE+gl9C8MDYYbI4XRxJhj3DCBmL2YVEw+pgJzA9OLeY6ZxnzHYrGsWDGsGrJ+3bEh 2P3YE9jz2HpsJ3YYO4ldweFwHDgpnDbOFkfCxeJScedwNbgO3BPcNG4NT43nwyviTfAe+Ah8Mj4f fxV/B/8E/xG/QUVPJUKlSWVL5Ue1jyqbqpyqneox1TTVBoGBIEbQJjgRQghHCAWEOkIv4S3hGzU1 tSC1BrU9dTB1EnUB9TXqe9QT1D+IjERJoiHRkxhHzCJWEjuJr4jfaGhoRGn0aDxoYmmyaKpoemhG adZomWhlac1p/WgP0xbRNtM+oV2ko6ITodOn20OXSJdP10T3mG6BnopelN6QnkR/iL6Ivo3+Bf0K AxODAoMtQzjDCYarDAMMM4w4RlFGY0Y/xhTGS4w9jJNMKCYhJkMmMtNRpnKmXqZpZiyzGLM5cwhz JnMt8yPmJRZGFmUWF5YEliKW2yzjrChWUVZz1jDWbNZG1hHWdTYeNn02f7Z0tjq2J2yr7Fzseuz+ 7Bns9ezP2dc5+DmMOUI5cjhaON5xojklOe0593KWcPZyLnAxc2lxkbkyuBq5XnPD3JLcDtz7uS9x P+Be4eHlMeWJ4jnH08OzwMvKq8cbwnuG9w7vLB8Tnw5fMN8Zvg6+OX4Wfn3+MP4C/rv8SwLcAmYC cQKlAo8ENgTFBJ0FkwXrBd8JEYTUhQKEzgh1Cy0J8wlbCx8QrhZ+LUIloi4SJHJWpF9kVVRM1FX0 uGiL6IwYu5i5WKJYtdhbcRpxXfFo8TLxZxJYCXWJUInzEkOSsKSKZJBkkeRjKVhKVSpY6rzUsDRG WkM6QrpM+oUMUUZfJl6mWmZCllXWSjZZtkV2UU5YzkMuR65f7pe8inyYfLn8GwVGBQuFZIV2ha+K kopkxSLFZ0o0SiZKh5Valb4oSyn7K5cov1RhUrFWOa7SrfJTVU2VolqnOqsmrOajVqz2Qp1Z3U79 hPo9DYyGgcZhjVsaPzRVNWM1GzU/a8lohWpd1ZrZJbbLf1f5rkltQW2Sdqn2uA6/jo/ORZ1xXQFd km6Z7ns9IT0/vQq9j/oS+iH6NfqLBvIGFIMbBquGmoYHDTuNUEamRhlGj4wZjZ2NC41HTQRNAk2q TZZMVUz3m3aaYcwszXLMXpjzmJPNq8yXLNQsDlrctSRaOloWWr63krSiWLVbw9YW1qet39qI2ETY tNgCW3Pb07bv7MTsou1u2mPt7eyL7D84KDgccOh3ZHL0drzq+N3JwCnb6Y2zuHOcc7cLnYunS5XL qquRa67ruJuc20G3QXdO92D3Vg+ch4tHhcfKbuPdebunPVU8Uz1HvMS8ErwG9nDuCdtz25vOm+Td 5IPxcfW56rNJsiWVkVZ8zX2LfZfIhuSz5Hk/Pb8zfrP+2v65/h8DtANyA2YCtQNPB84G6QblBy0E GwYXBn8JMQu5ELIaahtaGboV5hpWH44P9wlvi2CMCI24G8kbmRA5HCUVlRo1Hq0ZnRe9RLGkVMRA MV4xrbHMyCbvQZx43LG4iXid+KL4tb0ue5sSGBIiEh7sk9yXvu9jokni5f3o/eT93QcEDhw5MHFQ /2DpIeiQ76Huw0KHUw5PJ5kmXTlCOBJ65GGyfHJu8vJR16PtKTwpSSmTx0yPVafSplJSXxzXOn4h DZ0WnPYoXSn9XPqvDL+M+5nymfmZmyfIJ+6fVDhZcHIrKyDrUbZqdskp7KmIUyM5ujlXchlyE3Mn T1ufbj7DfybjzHKed95AvnL+hbOEs3FnxwusClrPCZ87dW6zMKjweZFBUX0xd3F68ep5v/NPSvRK 6i7wXMi8sH4x+OLLUtPS5jLRsvxL2Evxlz6Uu5T3X1a/XFXBWZFZ8bMyonL8isOVu1VqVVVXua9m V8PVcdWzNZ41Q7VGta11MnWl9az1mdfAtbhrcw0+DSONlo3dTepNdddFrhffYLqR0Qw172teaglq GW91bx1us2jrbtdqv3FT9mblLYFbRbdZbmffIdxJubPVkdix0hnVudAV2DXZ7d39pset59ld+7uP ei177/WZ9PX06/d33NO+d2tAc6Dtvvr9lkHVweYHKg9uPFR5eOOR6qPmx2qPW4c0htqHdw3feaL7 pOup0dO+Z+bPBp/bPB8ecR55+cLzxfhLv5czr8JefXkd/3rjTdJbzNuMd/Tv8ke5R8vGJMbqx1XH b08YTTx47/j+zSR5cn4qZmpzOuUDzYf8j3wfq2YUZ27NmswOze2em56Pmt9YSP3E8Kl4UXzx+me9 zw+W3Jamv1C+bH098Y3jW+Wy8nL3it3K6Pfw7xurGWsca1d+qP/oX3dd/7ixdxO3WfBT4mf7L8tf b7fCt7aiSBTSzl4AhdzhgAAAvlYitYA7UgMMAUCg/V0b7DAAQEEIB8FoZC8qDizBftAKwZAj1ABz wzkoFlQD2gcjiFnGjuDu40eoVqn5iCE03XTC9IWMQkw1LLqsy+xFnHbcNDwP+TIEnIW4hN+IXhL3 kxSRGpVJk+OTb1A0VRpTiVWjVr+oqaLVqW2iM6TnqD9kaGnUZSJvWmSOsgiw7LFms4m0bbVHO1g5 Zjs9dmF0dXBLc+/0WPEU8LLaE+ed73OL9I4M+Qn46wS4BoYHpQYXh9SHdoU9Dh+LWIxci4YpxBiO WJE4xXidveYJ7vuCE0P3mxzgOvD5YN+hssMHkjyOqCdzHgVHZ1KGj3WmNhyvSCtJL8w4m1l4ouhk WVZtdtup/pznudOnv51ZyZvLf332YUHXuRuFjUXXimvP15U0XLhx8VZpd9nApeHykcujFR8qZ698 qvp6daV6ueZr7ee6b9foGuQbbZpCrx+9UdLc3DLY+qZtvn3jFtVtljv8HRKdil063RY9bncDeuP7 0vtL7l0fGLzfMVj2IOmh5yP1x0yPF4d6hwufRD81fsb6bPb5zZGsF74vlV9hXr14XYOsJ+d30qPo 0edj+8f5xjsn/N5Tvb82uXsKN9U4TfpA9+HWx5AZ9pn+2dg5/rn783sXhBYefIpb5Fns/hy6xLzU +sUTWT3l3yy+fVnOXVFYefx9z/fZVdfV7jXONZ+1irWFHzHrXBt1v4S2tnbizwikgQOyC7wHsUBk 6A4sBpeixFAtSPz5MXPYu7jr+Faqh4Q5Ij9NEG0PvRDDSSY8cwYrD9tNDi8uFHc1rys/rUCX0AER CdFR8VxJJalqGTrZULkBBRnFbKUVFXfVDnVBjcOaY7t0tAt1vuvZ61cYrBuZGxeYvDcTM4+0qLac tRay2W2bY9dvv+Eo7uTknORyybXP7YMHaregp5aX+54Y7zSfYlKdbxf5qd+k/5eArSDaYO4Q8VDl MP1whwj/yNio9OhCSmPMYOxwXHN87t7QBLN94on4xLn9Dw80HCw6lHI4Imn3EYtkzaPSKbzHGFNx qZvHv6bNp09lvMt8fuLhyZ6s9uyrp4pysnKPnI47E5EXmO9z1rPA7ZxroUuRc7HzeecSpwvOF51L ncocLzmXO162rbCuNL9iVxV/taL6Uc16HV+92bXIhjONTU1Pr680s7QotTq0RbVn3Wy8NXJ7o0O4 06brQHdtz3gvsY+rn/ke7t76wOf704NvHww97H10+3HTUOVw0ZPspynPEp4Hj7i+MHqp9IrvNeH1 8puxtz3vcke9xsTHVsZ7Jk6995yUmPw+1TWd9sHmI9PHFzOFs+5zbHMj86cWLD9hPt1ejP4s8nl0 KfuL3pclJPpuy4Tl5hXf73Tfb646r35cS0fyxtb6p03qn4G/2reUtvZu3dyJPwyoASdQAHZINVAJ xpE9PQmqhSHYB76H0kK1IRXSS8whrAJ2DteKP0EVRvCkdiDa0jjT+tLtpT/N0MT4hGmFhYVVi82D /SBHMWc711PueZ4NPhw/m4CEoKaQtbC3SIRooliq+CmJc5IXpSqkq2SqZK/IlctfVMhTTFVKVA5W cVU1UpNVZ9EAGh8172td21Wkna6zTzdIz0Ff10DckMXwp9G08YBJrelps0RzLwtdSyErrNWs9aBN LbJm4u29HUwc5ZzYnFHOSy7Drpfd4tz1Peg9xnc3eB71ctojsueH9wOfElKMrymZi7zk1+tfEBAe qBfEGjQf3BWSFxoctiucGD4ccSJSN/JrVHm0IwVDaYkJjGWLvRe3L14qfnRvToJJwua+G4lR+6X2 zx6oPhh5SPHQj8PdSSePeCXLJP88+jCl6Fh4qs5xuuPjadfSUzJcMyUzf54YPlmetT/b4ZRkDirn bW7b6bwzlDzbfKWz3AX4gm/nJgqHirqLG8+XlxRdOH0xszS1LOVSannm5VMVeZXFVyqrmq52VD+s Ga/9Uo+5xtkg12je5IdkmbLmnpapNmK7wk33W8m3a++87sR0yXXv7sm429T7rp9wT3GAdL9mcOOh 4aPDj9uHPj0Rfur/rPz51xeuLwdeW71Zftcx1jrxYcr+o9Bc2GLIt/h19+34/z4j2v4mYFUByENq T5fHADhUAZCDnIOIUwBgIwBgRwOAkwaA34oD+MoKgGKU/vl+sCMVpjVSDx8C+aAOqRjfgWWICAlC mpAdFAQdhs5CdVAfNAatwfSwOGwAe8Lx8Cm4Gu6D38NbKC6kjnNGxaJOo5qQuu0bmgWthvZAH0Rf Qg+gvyKVmSkmBnMB8wDzAyuOdcUex7ZiZ3HcODuk0rqNW8ZL4wPwZfhRKm6q3VTFVGMEIUIw4Rrh B7UhdS71BFGeeIz4mkaeJp1milaf9iLtLzofuj56GfozDIAhimGC0YnxIZMRUwfzLuY2Fi2WDlZT 1idsXmzzyDpm4KjiNOQc5TrILcDdyxPGy8h7iy8QqSB6BWIFRQTfC/UJN4qUimaI7RMPlHCVNJKS l+aXoZb5Ljshd1++SaFY8ahSuLKzipaqoBpBbVl9VGNQ845W/6432ou6kB6HvoSBhqG1kbcxxeSY aYFZnXmvxTvLDWsOGxXbPXbH7K85vHZicDZzOeZ61+2D+9Zufk99r6A9ed6dPku+omQ/v3L/6UCp oPjg3lDWsMjw3kjOqNjowRjR2KS4F3vlE9L3Te6nHBQ69D7pcnJEilYqy/GV9OeZfSdbs+tzKk6X 5lWerT3XUNR6vudCT+lYeW7l7qvsNU/rkxqVr4+1nGt3us3cMdZ9q/fUvUODUY/2D8c9y3lx8/W7 UdSE+VTlTOqn/JW4H8rrLzY+bb76WfGrdCd/sCK5w3In/mfBNXAPOSlYgxggCUgfOQuIhjKgS9Bt aARaggmwMKwDu8OxSPRr4AH4AwqNEkTpobxRh1ElqE7UezQaLYa2QEei89Gd6M8YPow9JgXTgpnH 8mNdkKr7LnYTp4qLxtXhFvCS+BB8NX6RSoEqnuo2AUOwJRQRZqhVqdOo3xBlianEMRoNmnyaFVpn 2lY6TrojdPP0LvTdDMoMlYzcjHlM9EyZzETmTBZaljxWXtZqNg22e+y72Zc40jmFOO9weXL94i7l MeaZ5c3iU+Mb5c8UUBYYE8wU2iOsKyIuSiu6LDYq3i9xXbJYKk06XoYsay+nJy+nIKjIrESlDCmv qnxWnVObUZ/VWNRc3oXRZteR0tXTc9MPNThiWGjUYPzAZNoMMudDshjJKs26xmbEDm2v5ODjeNbp iQu1q7VbgnuKx+nd1Z53vN7u2fRhJWn6+pCz/G77zwVyBzkEZ4R0h0HhOhF7I5uilijiSJ4qjXu1 lzHBct+xxLEDvgdXD2cfEUvuSHE79uN4UbpWxsSJrKxd2Z9yLp52zePIny6oKzxU7FyieJG29Fe5 eIX3ldyrvbVQvUHD0aahZrnW3Jvw7X2duO4zvar9o/dzH3oP8T7ZfP7h5d03taNtE31TYzPZ868W i5c+fn22HLOytnp9J/4iyKlSArgI+sECxAipQp5QEnQZGoS+wOywLnKakw23wVMoepQOKgx1HvUI DSNveARyBjOJ7C7JmMuYGawMNgZ7E4fBOeBKcV/xxvgi/Dcqa6pqAjUhgvCMehd1DZGLmE2DpUmi +UWbRAfTZdCz0ZcxyDC0MZoyvmWKZqZmvsxizDLDmsOmxfaB/SyHJSeEfIfiudW413hu8h7kM+Kn 5h8RuCAYKmQjrC2iLCopJijOI8EtySPFKy0mIyurKWcu764QpZimVKHcozKlRlBX1iBrFmmtagfr jOv56U8bUowhk3wzafM+S2+rdZtsO3H7W46WTuMuka6/3I/tZvS8uEfde5BE9t3yKw3QC5wMzghV CnsfkRmlET0dUxBnHL+ScCXR6wD6YPlhi6TF5NwU5WNvjx9NF87oOxGaRZNdm2OWO3HmUD7v2Zvn nAu/FGeVSF3oLiWVbZTnVahWPqkKrMbUnKtTqh9sCG5cu57XLN1yt82rfflW5h3xjoGuwB7c3fI+ 2/4fAyWDVg/hR+eHuIZLnko+6xgJfIl/1fYm9B3taN2468TCZPjU1AfHj22zzHN280cXaj/1LQ5/ frbU9+XG11Pfwpall+dXSr6bff+6mrbGvVb/Q+7H1XXh9ZINeCNg4/6m7ObxzeGfXD/9f9b/XP9l /attS3Qrdzv+MQFKittfDwARDZDjx9GtrW+iAOByAfiZs7W1Uba19fMSUmy8BaAz7Pf/HbbJWOT8 uzhlG/236/8AleWgQatZQGAAAAAJcEhZcwAACxMAAAsTAQCanBgAACAASURBVHgB7d0PdBPXnS/w 660d5GTl8KeQ1ITaxJAFXhi3mDwgwSRSaIoPG6S2kHTBpKHsypy0D4tsgBXduGdNul7RbsA+Kcd2 NhFtbLd5ou8h2tTePGwTE0A+qUyQU+wldiOHIzeRE5FKKRJIW797ZyRbf2zLsqSxZH3nHJDmzp37 53NHP43uyJqM4eFhggUCEIAABFJK4K9SqrVoLAQgAAEIMAHEbhwHEIAABFJPALE79cYMLYYABCCA 2I1jAAIQgEDqCSB2p96YocUQgAAEELtxDEAAAhBIPQHE7tQbM7QYAhCAAGI3jgEIQAACqSeA2J16 Y4YWQwACEEDsxjEAAQhAIPUEELtTb8zQYghAAAKI3TgGIAABCKSeAGJ36o0ZWgwBCEAAsRvHAAQg AIHUE0DsTr0xQ4shAAEIIHbjGIAABCCQegKI3ak3ZmgxBCAAAcRuHAMQgAAEUk8AsTv1xgwthgAE IJAJgpQW6D1dp79sW/TQU0/L82hH+lvqmt6xLXhoe5m8gK4OtJ34+flrC1ZtlP7+TD8peGrv9jwJ Ic6B002Gyzb3Q9u3ffSrhtF0HsI72PniK2/enFXw1LPb8/ijwz3QVvPz82TBQ3vL5HTvMRdaUd2Z axt375cXjJdlzP2CEllFdWcWbNz9NN/4oG3jr0xK4IHtZZsYSNDiHTjx4s+vURZ/T4O28iveoc7j tW+6Z83yb5IsuCd/7dceWzZ/6t30FzX+o3ew6cVXQsZl/NwTbBmsUZaUk90DL685FdQLQiQLHlZ8 a02edIKdJ7vJPdj2mwtZ9z9avGzOUGfNxrXlWw19z28J055sccg3aQF6r2EsqStg1pWyoVboHKwP Ni3HD7ysll911CrYquqVk0JyB0t16fhEmq5tezMg3WfgMFULRRj5ImiqL0VWbfdlGePBVM0KrTZN kGWMvUKShIpk1aaQ9IlXJyXQ2DdGIQ4TLyEb6Wl4Hr8GTzL6n6LDFp43fikOY/i4TKF0S7OGNllR bRqnF6Sxxz/GUyjdt4ungx96TmtkCS4TfzhqLFMvEHtOVgBzJqOvyFR8trS4hDXbcNHqJeR6b4uZ 70T72+87CXFb3jbQVW7zo5vPWi0Wq20dPc3yWi6yRFmzxXVAJhtN5/dj/2UJ55g5WaEps4I/o3nd Xlqlb8maFdMZnJcvKvv2HFoc+xfNMimBh9iHktAli/CNXnh7Nt0iNCE0i1+Dq201mc0mQ62az2G4 9IfrYVnjlyAtGmNcoi7e2d5YRXd64rEVwb0wG1t9b99a/btRlzq6g7f7dM1OedaGcnY8LZbwx4tk hUJF16rae+nxhyWxAojdifVNdOmSvDV8OKl/z0rcH1xq99XX8I7FS6zdDWx16+q7rC//yz+XH9T1 fdZ76LHl9Syx/eCubzdd/r0v3c2SJrG4Ww6XKZU7yw7tU2ZkZWdlyctqwl+k7sHOI2XyDH5RltV0 DflCfG/biZ3yQiFdvvPwuQG+1uvdR2gAyMrKKNz5ov7i+G1wt9UdPtLUOfp24c86KYG8TOfAuUM7 hVYV7qtpCQi9DUf2lxVmsCbsqzs3jgT3sHzVypWrtpRVVPKnxDSbu/ekUi4/VHe66bAyI0PZ6SRj dNw7ULOT5iprGaANd5+mLZArT3SxynubDsnp7k29/F60BLYUKve19F4n7oGRcaG17JQr9x2uOUzb yBblic5BvuvOtrp9fEpG2eFDZTuVOw+dDIqXbksrG371miUj0zvcgxtoL1aukT/9b/zHNfPVfqe7 97CSDumRfr7nvScPywuVTb00ebx6/e7E2awob/AfcP5UyVe/xo7HU29e8afgMWECkz1BR74kFfAY 1CycaJr7TMIUCX+olDb29BnYq4jTtNJZD35yQGG09WlLZcKhJCtVGy53+NIDPjo7zLV8BoXJn+hL YfMwjlrf3kShViv4KEZk1XT+wFxbSvdicyZ2I18m4UpHcpSaXf6JF8JpdY1aNV8KK9A/yUNzq4T9 2Gf8saT5qrlqf6MCs0QW8FibhcaWqjUqGXtKWTy+D/isu6X+2lWNPYFF+zWISqOlC92Z5SaVVjqV 5IPiE4jC9OGYHadtYxlK2aRNDzNiVXfQvav5kmpNHwrnwDJNrV5XKWw3DfrHi4qP1CIr9Q+dqscz 3CPMlRFOpVYJXSOy2sAZK1dPIyuNnz3zFyIz9FhtVqu1z1gpDHytefTY4GXN/CFEx9G/Cy0hqN5A HFtfT09fj17DejIyar4dgxsTuBeex0uAxKsglDNdAhY+PMjU1fwLUmHoaKQvJk5VXcuHyEo6NesL Unw49pj5CKIwe0ZmJ0fDNO2C/0U7mjjyaqSRRAg0pbV8eHUIk5tcs9UzEruFJ5xK72IcDiHCqPR9 Lqu52aA3dPS4XA5TI5uHpWHFJsQXoqLBnS7Gata0kSjAkujicthsNru9jwU7rrLHbqerDtr4gCWi gBCSZBq9xWa3mHR8sFOYbEL76bQvq154q6NvRYER0K/B2huwyPQ9Dv8mWaOZTX6P13F7h5btWKq3 9ul9JXBai62DD93qPo/v7VCmrjWaTc16favRbHcKDWND4MfXslY5jPxeCqPDt5eWn3d3mHWsZIVw kcPnIuzIaZoplb+pvvr9D2p2ESDw2PD3YjR2y0Lq9RUe+CC8i4yOmtC14MYE5sfzeAlgzsR/JKfs 48LV7Eyy/Vh5BZ145B4sLn5kLSHm+vI9x+gHWtkj988P6pnLI6zecAUlj674tkuJf8Lbe+Mm25oz mmX9g0vZinRFCQu2ZvO1oA/rbNM9d/Ef1KVrH2U53h/8U2Y2ufpGg2LD8uzsnKIdbB52tEDFuqV8 7kLZepYevHTVrF+wYMHcuUvKaW/MFcvnzqWrx0zXA3NFEpD+V9cHNH971bb8BXPzi3YJFwV8HSWl K/NZ9QXFj7PImBMyrS/UozD6IrqjmZ1mtm872v7fwhbVM0+sDBAO6/icB3ikBkNDExseFT15Nrf8 vJZNNnCV3yzIlD5RracfBdqP7VnLFZVs2/bCqxccXqHo0f9lmx6eQ9ek927igzcdmVlsOBQbH2BV S5cXMeXgJYu/brE4b2HAVQpOq29ubW7W1/LvneTYr84J0y8BewoXO/wJ4fX6t4w+evijY3RdOnvh 6AqeJVAAsTuBuOIUnZlbuNlfk2zrujkkdwM/i8LSZMr72Ys+YPFHZP9jwCb+qXTxCj4+NJy5MMAn eDvfPMWeOG6OhJSubiu/yfoeP6F+zxfZxT7/wsdx15/5Ve97l1mcXJiTZX6tvLzeoNA22zzDduGr LA665RbLZrjMLrQSYunuYg/By9Indc3Nza2twsxQqb6VBp/WbfcFXRqNJCBZ9D8W01IV1UbPsMfl snUY9I36vYt8rTZ/YGdVDl06T+NpYDdZqn/J8oVAP9v7FuH9SrHqb/zBceyOE8nybWw4Gg5WNBDF MwcO/D0N/RUV7KLDM5u/QrxDV61//Y0DL1npJwujns5ktNfved3MaAKXHOFKIP1mHwvZbOEDpqHu l0ys/7en2DgE7+ThY6rh7W5+HpvtQt8sSpSb5Js2bS37145KNsgtlz4iHuJ742U983bzc+QsL7+E 1+vfMu6j1/4BG3IHGTlaxs2KDbEJ+A+82ErB3tMqkLdRI6uqYpFH9sgy+j/39c3kGHsFKZTrhdAt vD6FM83xnvu6MOf+JxWk3UAOPprfqFDM+8DQzkoiavXmkXeB+h3Lb/+kNvud4/z5s+ahAonjjG/v 5Y/v4fYYzFUlO7OrH3C1lbNmyFSKleQ1FnUcfxzovXi6+YflLHcOkSyT0TPAKnJsydbsWrlrTzl/ GdVXku9BmrdqUx597v6Qht95j/6tfMzvmEcQWCFT0rcIQ7mmgjx190enyqsMdOLd9rcb+TrMioU7 tdpFjQdZb0bEfNX7HgxF6wtpAKafMXgMovr7zbMJ3+ebvtP3cTtOyLpv7hCGQ/bgVwpWfpF+EYPv p3ojR9+BbK+WlNDVUq1etfqWUJvk9i8EjlFwS4Q16abvaoihqn5XUf0u/3Z/WBfWswh/Cm11+j9e 0GQn+7DFv+vNXsCfHP/pzyQ7l72tEcNz/3Bo06IrB9mbQEyL69M/MqKl8wLfz2MqETuPJxCvyReU M40CNmFSlSg6hI/21mZ2WkVIrXDB0SXMcZeytfGeB7be0Ver5i9m+Q4arlJv4meYffPdRDaytdTQ xy5yCbO9tfz3u22mxpHN9JKl3sza5LG2lvpKo3sL29V0ntnV1zySzrHoSKdtx75WyabaS4WvsQe2 1fc8ggCdzm6u5ovnG6HQGOn5vzDVy8lKfVdd6ZuMLuR7264enb/VvkdOptA2GukEuTCJHNjaMTvO 2uczJ9Vs5sWj5y8uyipbhaY7egy+K6B8DeraDlfAGAXXIvjz4zjsMdNvLKpUpapKvUHHhlsRdK1y 2NPHXyWlk+PD/l4IO7JqHSb+ijSnpRdd+wzCNVJahEyQoOM4fr1Cq0f/F75fX0ove/KLsBpyyXc0 N57FTwDXKuNnObNK8jjYVUG6uEYvDPpidy0NunSz3T66JbTv/GY7/92F0U0s0RFQnH+LhxZlD7n+ 6N8W10cXXxF/YTS4XNaZ+DRgzI4HVzbWmssxjs1YmWmaMOmh1tFZoGFLM3851PcHWaM7GLUspLOL 1REXFwMYY2Qi7hiawc5/f4ZrnkSdobtiPUqBDJqff7/HfxCIKOCsKcwpNxOt0X5gzcgMSsS9kCH+ As7ek+uXbxMmcITSdSb706uCB2WwJWNhCeGq7Zf3Bm+If3uEEp3ddTncHqI2eI5uwWxsopT95SJ2 +yXwOAmBof7ua38idy9dnivFa3MSXgnN4hzsfMd8zf75bX+9qPChoryxRmSwt+sjz51LVxYEXdtN WKu81wfMVz+9e8WqXHHqS1hHUqJgxO6UGCY0EgIQgECQAL4jGMSBFQhAAAIpIYDYnRLDhEZCAAIQ CBJA7A7iwAoEIACBlBBA7E6JYUIjIQABCAQJIHYHcWAFAhCAQEoIIHanxDChkRCAAASCBBC7gziw AgEIQCAlBCLGbm9X02H+jh3ywye7vHyf+ltqhBuQHPLfxyQ8xdnfItzoQ3moaZDfLTwlJYDQSAhA AAJJKBDhr+OGzlUV7TjZaOxbQzqXrC2a1WF79stvLSkpr27t0c1+b0vR2uxFNk1YyvPrHM8tKfm4 utWimF2XX1Qye9HlZ3NDUw4UJyEHmgQBCEAgJQQi/F1l15HCIqIbPrCKdqalLONggbHhTg3XdWC4 bhNNYVtv1psXhKY4lO/mcL3W4aO59Hcnu47kFJEPzXd+OTjFPnxAnN9YSIlhQCMhAAEIRCUQ4bx7 1bOdnkx2VxHvQAv9meHanhWeNwlXsECoY9G6reT7nZ/tDk35nP5aMlks/KSBdNFqQl7o+zNNCkq5 6jywZhI/eqDX++8UFVW3kBkCEIBA0gts27Ztym2MELtJpiSTuLuaKul9qlQ6U9kyafebxExvtsEv kjtmEbObTpmHpHgJDfduYXKcSO6QkU/57OEpkZsdS98il44cEIAABFJTIOK1yqE6ZXbRDleHxVH3 NJs5YWH7M98NPixvtxDNanaDjOCU2SyT7xZZbktnO9maf1toyr2TOOmm+2CBAAQgAIFwgQjz3b0n lMt3FfU5NHlZLpeLZEmlHnNNTtGpDuubxdK+fTnLbxgsP7nHEJJSJ/tUnlO0qcN6oFh6cl/OthsG z0/ueSwkpQ6/8Bs+HEiBAAQgMDmBCe/V4L/Hlb+oanZTK1er1ndPK07VyN8fIzxl2NLK38uD7ajm 74Q1RsqEVWMjBCAAAQiMKxDhvNsftEMfvW4nOwuXsMuYwhKeQvgkqXQ0zxgp/t3xCAEIQAACkxeY YuyefAXICQEIQAACcReIeK0y7jWiQAhAAAIQiFUAsTtWQewPAQhAQHwBxG7xzVEjBCAAgVgFELtj FcT+EIAABMQXQOwW3xw1QgACEIhVALE7VkHsDwEIQEB8AcRu8c1RIwQgAIFYBRC7YxXE/hCAAATE F0DsFt8cNUIAAhCIVQCxO1ZB7A8BCEBAfAHEbvHNUSMEIACBWAUQu2MVxP4QgAAExBdA7BbfHDVC AAIQiFUAsTtWQewPAQhAQHwBxG7xzVEjBCAAgVgFELtjFcT+EIAABMQXQOwW3xw1QgACEIhVALE7 VkHsDwEIQEB8AcRu8c1RIwQgAIFYBRC7YxXE/hCAAATEF0DsFt8cNUIAAhCIVQCxO1ZB7A8BCEBA fAHEbvHNUSMEIACBWAUQu2MVxP4QgAAExBdA7BbfHDVCAAIQiFUAsTtWQewPAQhAQHwBxG7xzVEj BCAAgVgFELtjFcT+EIAABMQXQOwW3xw1QgACEIhVALE7VkHsDwEIQEB8AcRu8c1RIwQgAIFYBRC7 YxXE/hCAAATEF0DsFt8cNUIAAhCIVQCxO1ZB7A8BCEBAfAHEbvHNUSMEIACBWAUQu2MVxP4QgAAE xBdA7BbfHDVCAAIQiFUAsTtWQewPAQhAQHwBxG7xzVEjBCAAgVgFELtjFcT+EIAABMQXQOwW3xw1 QgACEIhVALE7VkHsDwEIQEB8AcRu8c1RIwQgAIFYBRC7YxXE/hCAAATEF0DsFt8cNUIAAhCIVQCx O1ZB7A8BCEBAfIFM8atEjTNe4Cvf+1m0fXz3p9+Jdhfkh0A6C+C8O51HH32HAARSVQCxO1VHDu2G AATSWQCxO51HH32HAARSVQCxO1VHDu2GAATSWQCxO51HH32HAARSVQCxO1VHDu2GAATSWQCxO51H H32HAARSVQCxO1VHDu2GAATSWQCxO51HH32HAARSVQCxO1VHDu2GAATSWQCxO51HH32HAARSVQCx O1VHDu2GAATSWQCxO51HH32HAARSVQCxO1VHDu2GAATSWQCxO51HH32HAARSVQCxO1VHDu2GAATS WQCxO51HH32HAARSVWBysdvdvTNjZ5db6KT79OGdcrYoCwvlTb1OmtrfUiPPYMuhpk4vn8vZ31JW yFKUh5oG+aTwlFQ1Q7shAAEITLdApHueeYdaGn7ReLS8gSi+7yFEQojXoq9oeMrYU3yn58YNMm9h tnfg5JKS8urWHt3s97YUrc1eZHt+neO5JSUfV7daFLPr8otKZi+6/GxuaMqB4unuO+qHAAQgkKoC GcPDwxO1nQbmV9pvzbq2Y5fJ6Di1RkrI4OnChcbXbd/PvpG1MG8+jf3ddXKu68Bw3SZaTteRwqKb 9Q7luzlcr3X4aC4hzq4jOUXkQ/OdXw5OsQ8fmDNRxdiWwgK4X2UKDx6aniICkc67M/O2lj1NvN3/ satd6NH1a71mUrV8wRuEmAlRGe0/zbpJuIIFwtZF67aS73d+rqRri2mcp4t00WpCXuj7M00KSrnq PMDeCSIter0+UhZsnwkCGOiZMIroQ5QC27Zti3KP0eyRYreQ00unS/zLneuaO/o2FhdkkqE65QLV a2UNLIr7MkjumEXMbi+bW6H/84vkDhn5lH8WnuIvc/zHWPo2fqnYkliBH52N+j7xGOjEDglKn3EC k7tWGdDtzJzc5ffTwE2X+Q+XyMxuDwvbn90SsljebiGa1bNZ0k0hdrstne1ka/5toSn3TuKkWygT /0MAAhCAQIhA1LH7/YYl+XOPDLiJ93p33Z720nu+uHS9klT98Bz9Nomz99XydtXaAunS9TJSoTs3 SKe7f/NqOVEVLgpLwWR3yEhgFQIQgMDkBSY3Z8LKy8niS121t6+6e0l+9kG6xqkbz24vkBJVq7Zt w0K2nVM1ntmSR0ierlWbv2Ehy0TUJvuWTCkJTWGbsEAAAhCAwFQEIn3PZJwyvW63i2RKJaOh3+t2 ukiWVEJnuv0LnySVTpjiz4vHmSSA75nMpNFEX5JTYDT4RtW+TIkkZL46UyINSSHhSeEpUdWKzBCA AAQgwAtEPd8NNwhAAAIQmHaBKZ53T3u7o/1U/u5PvzPtbUYDIAABCMRLAOfd8ZJEORCAAATEE0Ds Fs8aNUEAAhCIlwBid7wkUQ4EIAAB8QQQu8WzRk0QgAAE4iWA2B0vSZQDAQhAQDwBxG7xrFETBCAA gXgJIHbHSxLlQAACEBBPALFbPGvUBAEIQCBeAojd8ZJEORCAAATEE0DsFs8aNUEAAhCIlwBid7wk UQ4EIAAB8QQQu8WzRk0QgAAE4iWA2B0vSZQDAQhAQDwBxG7xrFETBCAAgXgJIHbHSxLlQAACEBBP ALFbPGvUBAEIQCBeAojd8ZJEORCAAATEE0DsFs8aNUEAAhCIlwBid7wkUQ4EIAAB8QQQu8WzRk0Q gAAE4iWA2B0vSZQDAQhAQDwBxG7xrFETBCAAgXgJIHbHSxLlQAACEBBPALFbPGvUBAEIQCBeAojd 8ZJEORCAAATEE0DsFs8aNUEAAhCIl0BmvApCORCIXWCX/ttRFaLb9suo8iMzBGaMAM67Z8xQoiMQ gEAaCSB2p9Fgo6sQgMCMEUDsnjFDiY5AAAJpJIDYnUaDja5CAAIzRgCxe8YMJToCAQikkQBidxoN NroKAQjMGAHE7hkzlOgIBCCQRgKI3Wk02OgqBCAwYwQQu2fMUKIjEIBAGgkgdqfRYKOrEIDAjBFA 7J4xQ4mOQAACaSSA2J1Gg42uQgACM0YAsXvGDCU6AgEIpJEAYncaDTa6CgEIzBgB/AbsjBlKdAQC kxX4yvd+Ntms/nzv/vQ7/qd4TAoBnHcnxTCgERCAAASiEkDsjooLmSEAAQgkhQBid1IMAxoBAQhA ICoBxO6ouJAZAhCAQFIIIHYnxTCgERCAAASiEkDsjooLmSEAAQgkhQBid1IMAxoBAQhAICoBxO6o uJAZAhCAQFIIIHYnxTCgERCAAASiEkDsjooLmSEAAQgkhQBid1IMAxoBAQhAICoBxO6ouJAZAhCA QFIIIHYnxTCgERCAAASiEkDsjooLmSEAAQgkhQBid1IMAxoBAQhAICoBxO6ouJAZAhCAQFIIIHYn xTCgERCAAASiEkDsjooLmSEAAQgkhQBid1IMAxoBAQhAICoBxO6ouJAZAhCAQFIITC52u7t3Zuzs cvta3N9SI89gy6GmTi+fFp7i7G8pK2R5lIeaBvlM4SlJAYBGQAACEEhBgUix2zvUcqJm5xqugTg9 HtY/78DJJSXlytYei0n/xo61VeeGwlOIt/+5JSUf7261WEwrqnaUvHhujJQUxEKTIQABCCSJQGak dtz4/GZOyb7Khl0mIWdPy3Giat4rX0bIMp2WKzr7B+WC0BT17Hfridq6V55LiMakrSq6eK3kzpCU 6weK50SqG9shAAEIQGBMgUjn3Zl5W8ue3l6qlBGHsL/nJuEKFgjPF63bSk52fhaW8jm5SchiKZ9J umg1IS19fw5Nueocsz1IhAAEIACByAIRz7v5Irz8dAn/NIsQM/GtSu6YRcxuGv5DUrxEQohbmAon kjtk5FN+1/CUyO3T6/WRM00iR7zKmURVU8nyo7M3ot3tB4/cHu0uyZx/agM0tb2S2SFp2wbqRAzN tm3bplzs5GJ3QPEsbH92S0iwvN1CNP+cTVpCUmZ73iPkphC73ZbOdrL1ldvoHkEp9wqn5QElj/l0 vL796OzPxsw/XuJ45YyXX+T0aLtDm5fMPZpyd36r/1VU8smMEFVHRM485QESuZ2obgKBSHMmYbsu Xa8kVT88R7874ux9tbxdtbYgPEW6dL2MVOjODRLi/M2r5URVuCgsBZPdYbRIgAAEIDBZgcmfd+fQ 2RK6SFepWrVtGxayNU7VeGZLnpSEphCSp2vV5m9YeJDtoTbZt2RKSWgK24QFAhCAAASmIjC52C1Z 1TZ8yl+8RH7glGev00WypBI6r02X8BSSJz8wzGeSSoU8Y6T4C8QjBCAAAQhEJzC52B1WZqZEGjJf HZ5CwpPCU8JKRgIEIAABCEQUiHq+O2KJyAABCEAAAokWQOxOtDDKhwAEIBB/AcTu+JuiRAhAAAKJ FkDsTrQwyocABCAQfwHE7vibokQIQAACiRZA7E60MMqHAAQgEH8BxO74m6JECEAAAokWQOxOtDDK hwAEIBB/AcTu+JuiRAhAAAKJFkDsTrQwyocABCAQfwHE7vibokQIQAACiRZA7E60MMqHAAQgEH8B xO74m6JECEAAAokWQOxOtDDKhwAEIBB/AcTu+JuiRAhAAAKJFkDsTrQwyocABCAQfwHE7vibokQI QAACiRZA7E60MMqHAAQgEH+BKd7zLP4NQYkQmKrAjVProtr1duXFqPIjMwSSUADn3Uk4KGgSBCAA gQgCiN0RgLAZAhCAQBIKIHYn4aCgSRCAAAQiCCB2RwDCZghAAAJJKIDYnYSDgiZBAAIQiCCA2B0B CJshAAEIJKEAYncSDgqaBAEIQCCCAGJ3BCBshgAEIJCEAojdSTgoaBIEIACBCAKI3RGAsBkCEIBA EgogdifhoKBJEIAABCII4PdMIgDNvM3R/voHFcAPgMy8wwA9SnUBnHen+gii/RCAQDoKIHan46ij zxCAQKoLIHan+gii/RCAQDoKIHan46ijzxCAQKoLIHan+gii/RCAQDoKIHan46ijzxCAQKoLIHan +gii/RCAQDoKIHan46ijzxCAQKoLIHan+gii/RCAQDoK4O8qU3jUd+m/HW3rddt+Ge0uyA8BCCSh AM67k3BQ0CQIQAACEQQQuyMAYTMEIACBJBRA7E7CQUGTIAABCEQQQOyOAITNEIAABJJQALE7CQcF TYIABCAQQQCxOwIQNkMAAhBIQgHE7iQcFDQJAhCAQAQBxO4IQNgMAQhAIAkFELuTcFDQJAhAAAIR BBC7IwBhMwQgAIEkFEDsTsJBQZMgAAEIRBBA7I4AhM0QgAAEklAAsTsJBwVNggAEIBBBALE7AhA2 QwACEEhCAcTuJBwUNAkCEIBABAH8fncEIGyGAASSd61nGQAAIABJREFUXOAr3/tZtC1896ffiXaX ZMuP8+5kGxG0BwIQgEBkAcTuyEbIAQEIQCDZBBC7k21E0B4IQAACkQUQuyMbIQcEIACBZBNA7E62 EUF7IAABCEQWQOyObIQcEIAABJJNAN8RTLYRQXsgkKQCu/TfjrZlum2/jHYX5J+kQHrF7mgPPhx5 kzyMkA0CEBBZYApzJu7Th3fK2aIsLJQ39Tppi/tbauQZbDnU1Onle+DsbykrZCnKQ02DfFJ4ishd RXUQgAAEZoxA9OfdXou+ouEpY0/xnZ4bN8i8hdnegZNLSsqrW3t0s9/bUrQ2e5Ht+XWO55aUfFzd alHMrssvKpm96PKzuaEpB4pnDCI6AgEIQEBkgehjt+2qmWh+cG9O5o2s5cvm0/27m44TVfNe+TJC lum0XNHZP6hnv1tP1Na98lxCNCZtVdHFayV3hqRcP1A8R+S+ojoITFUg2r+6ngF/cj1VKuwnkkDU sfv6tV4zqVq+4A1CzISojPafZt0kXMECob2L1m0l3+/8XEnXFkv5JOmi1YS80PdnmhSUctV5YI2Q Y8Ke6vX6CbdPduPUypnaXpNtU2z5ptY2utfmrKgrnlpdUVUztSqEvaLt0dTqEqE7UVUhcuapoU1t LxG6liQN27Zt25Q7G3XsJneua+7o21hckEmG6pQLVK+VNbAo7hFaILljFjG7vURCCP2fXyR3yMin /LPwFCHHRP+P17cfnY3u12eEcn6r/9VElYVtG6/2sIyxJkTbHVofbVu03RH2unHqxWibG63D1LpD WxVtj4SGRdujaLtDGxZtj6ZQRbSDEkv+aLtD66I9inZ0hL1iaeck951adyZZeNJmi/paZWZO7vL7 aeCmy/yHS2Rmt4eF7c9uCT20vN1CNKtns6SbQux2Wzrbydb820JT7p3ESbdQJv6HAAQgAIEQgahj 9/sNS/LnHhlwE+/17ro97aX3fHHpeiWp+uE5+m0SZ++r5e2qtQXSpetlpEJ3bpAQ529eLSeqwkVh KZjsDhkJrEIAAhCYvEDUcyar9vZVdy/Jzz5I6+DUjWe3F0iJqlXbtmEhm0blVI1ntuQRkqdr1eZv WMgyEbXJviVTSkJT2CYsEIDADBe4cWpdtD28XXkx2l3SMH/UsZtICva+NvzMy24XyZRKhN0l8gOn PHudLpIlldCZbrbkyQ8M80lS6bgpQk78DwEIQAAC0QpEH7v5GjIlkpD56kyJNCSFhCeFp0TbXuSH AAQgAAFCop7vBhoEIAABCEy7AGL3tA8BGgABCEAgagHE7qjJsAMEIACBaRdA7J72IUADIAABCEQt gNgdNRl2gAAEIDDtAojd0z4EaAAEIACBqAUQu6Mmww4QgAAEpl0AsXvahwANgAAEIBC1wBT/Nifq erADBNJPAPfYS78xF6/HOO8Wzxo1QQACEIiXAGJ3vCRRDgQgAAHxBBC7xbNGTRCAAATiJYDYHS9J lAMBCEBAPAFcq4xsjR8gjmyEHBCAgLgCiN3ieqM2CEQSwLlCJCFsZwKYM8FxAAEIQCD1BBC7U2/M 0GIIQAACiN04BiAAAQikngBid+qNGVoMAQhAALEbxwAEIACB1BNA7E69MUOLIQABCCB24xiAAAQg kHoCiN2pN2ZoMQQgAAHEbhwDEIAABFJPALE79cYMLYYABCCA2I1jAAIQgEDqCeD3TFJvzNBiCEAg LgIpfWMjnHfH5RhAIRCAAAREFUDsFpUblUEAAhCIiwBid1wYUQgEIAABUQUQu0XlRmUQgAAE4iKA 2B0XRhQCAQhAQFQBxG5RuVEZBCAAgbgI4DuCcWFEIRCAQLoIRHtTutuVFxNBg/PuRKiiTAhAAAKJ FUDsTqwvSocABCCQCAHE7kSookwIQAACiRVA7E6sL0qHAAQgkAgBxO5EqKJMCEAAAokVQOxOrC9K hwAEIJAIAcTuRKiiTAhAAAKJFUDsTqwvSocABCCQCAHE7kSookwIQAACiRVA7E6sL0qHAAQgkAgB xO5EqKJMCEAAAokVQOxOrC9KhwAEIJAIAcTuRKiiTAhAAAKJFUDsTqwvSocABCCQCAHE7kSookwI QAACiRVA7E6sL0qHAAQgkAgBxO5EqKJMCEAAAokVQOxOrC9KhwAEIJAIAcTuRKiiTAhAAAKJFUDs TqwvSocABCCQCAHE7kSookwIQAACiRVA7E6sL0qHAAQgkAgBxO5EqKJMCEAAAokVQOxOrC9KhwAE IJAIAcTuRKiiTAhAAAKJFUDsTqwvSocABCCQCAHE7kSookwIQAACiRVA7E6sL0qHAAQgkAgBxO5E qKJMCEAAAokVQOxOrC9KhwAEIJAIAfFit7O/pawwgy7KQ02D3kT0BWVCAAIQSBcBsWK3t/+5JSUf 7261WEwrqnaUvHguXYDRTwhAAAIJEBApdjt7ztQT9fG98ry8VRqT1nzw4vUEdAZFQgACEEgXgWFR Foe5mpBqh1CXrZUQmdG3EqH6dBkG9BMCEEhLgQgRcPzNmSJyuX2z3JI7ZOTTSdZLWz7JnMgGAQhA IH0ERJozIR5KelOI3W5LZzvZeq80fZDRUwhAAAJxFhApdkuXrpeRCt25QUKcv3m1nKgK58S5IygO AhCAQBoJZIg2KTHQdiT/0YM8rdpkP7oKwTuNDjN0FQIQiLOAeLGbNdzrdrqIVCqJcydQHAQgAIE0 ExA3dqcZLroLAQhAIEECIs13J6j1KBYCEIBAegqkcux2dsozCtuGhIEbPFyYIT/S6VtpOZRRWOcM GVJ3lzxD2RWaGpzJ2VmYUdjpJN6h7pOnu9zBG8dZc5/YWVjXPXG54+wanNxVI8847PuL08G2w/Tn A875/oRp8FBGRk1XyN8zOevkNDFCvV1HCjMOd9L+tJ1sifhTBN6hzsNK9sMF8p1HuoZi/eECZ1dN RsZh//i00HKPdApr3rZDGfKaruDeE2d3zRijFpzJ2XWEFkP7PNTdcrqTXvqecPEONh1Sst9hkJed 7PY1ZMIdJtyYLMcbcfeeKJSHHd4Ttn3MjdN+vLn7T++UC0uhfN/JCIfymH0ISJz+442Q7pM0DtGl 8HBTV6yvn4Cujf10/K9+J/8WWyUhmlYbbajHYmDd47RsZdjTrCZctSm0Ay6TjMjMntDkoHWPzdhq tHuGXWYtIVpX0LbwFY+5uVFTytGaq0328M3Rptg6aIcqhS4Y1KxDlR38mpX2TmYK/WsmRzVHqs0R 2mjvMRr77MMeEy3NGCGvjfaZqBv7rH2NtHYfZrSdCMjP/giL8OMzbOH7w1V28JsttPhwMYepmnC1 E7fRY+/pMPbRMWTjUxk2xAGV06emahnhKs1Wq5H1RxGJKnjnMdam/Xjj2+TpUVFWWW3o4TBGgyMk TffxNtyjUxBVY09fj5kuPdYIzY24ebqPN3sHPShlhh6r1dRIh6ixb+JjOWJ/ImQgEbYn92ZjJUeD N21jX6OKU1UqCNfMDgAWGmrNDptJr6CENAqpai2U0WWiq5xCxqfJajssNKvNrOdjLzVXd1hpzO7R KDTvvtvI4jFdODV9wYeWM2riMup1ukZd6ViRaDTX5J/ZO2i9fLCjr09ZpUbBqZvp3lb6XsReqza9 xteh6lbaeEct7YpMITRVXdvBjhSPTV9Jm8MWda2RxrieRo36xBsUhF+4xh7HsN1EpdjCqVqZi39x GDnC+eK7VU83T/JvX/37hz+yNwP+zdXTWErHh7ZCw8aHvRUpTC6PSU/fq9iiqm6l7XCYa+lzhdAh Nhw0zWPW0zFmi0yt48ensVSt/x2LxWzhVI1hLCPNcOhKZbVCwHZQWK4j5mg33ccb7ZpHryKlWq1C UR2Hk4XpPd7YORanNvTYrBabY+JTqpExnfjJ9B5vrkYFURsswx6H3eGwWfqs8enUuF1O7dhtN9Lg oHUMexpl9AzU2qwiqmbbsL2ZnWR9yOKgzmj1uGyN9ESFhnghdmv0NofDbKBRg2u1Xa/liKLWRGN2 q1ZGSvUedm5OY5arx6Ch+5itdo8trJxQTA8tRBuP8+7hYTs7m6Qh06KjZ6A2FuPU9MS7VUUUup4O GsRKdVaXx2am7+q08Q561kI4jdnmsJkNtLOVrVZ26kpKzXaPy9JKA7ve4jJVc5zWaOvR0+ClN7O9 aRWlOqOLdpW5CKf5vi65XMJLyNVMey/TBcT10D5Pct2o5Wh/6DuIgn5usPfRiGtg40PfinQfsg8Z KqPV4bKZWFCn/eFjt0ZvdjisBtpZUmllbydEZ7YPuyy0pFJ9H9/BarvLZtBwnMZgtbvCWEKa5mqt 5gO9Sh97eJj2441/F290ecwKThuH2D3Nxxs72uki48+mVLURPkWFjOuYq9N6vPHnUnyP2H+qxngM 0Ji99CX+1UhdqfhkzoqHOdJyZcDc3C5bvzy36Buq+v9r6r/UQRRPfGnokpmQq50nj9f/4r2PCak6 QydHHYSr12ydL5Wu3KLWEvOZXueXNnOGPUXfO1x/a91PbC9v5X8iYB4hkvzFdxNyd37uHNe1McoJ tnLdDF6PYW3Ow1ruZOeV7vOnuN0Pzs9drSbHLvb3nq8nyuIvXjlpJs6rJ+uP/6L9PULMvzZ9dPMD Ulm/f+V86fyVW3SVpOJMf/a8fI40cN/6Xv35Wz/ps23No1/HpN3Jmp9/7zwy796luRLXtRb691FX 36p/sf7tfupSYQqYBZdIMr2DXYcKs0uqVKZflcb+Xc77Ht5BTnb29xoN3JMr5hR8XUXeMPVe6jim eGrt0KV2Qj7uPKmr/8Xbn9Dx+TWd1aGQWs3WlVJp7pb99bRtvZ9/cTMhu7hvHa4/v07X9/LWApLF dDMl8xffPY/Mvid3jieMhY5z4JKZt/6b+moVqT98qn9y1y8C9w5+Pr3Hm3ewpaTkA9Nvt0vYnylL 4vFzFtN6vHm965qbe+yetjb6UbC2fs9z9DpTjMv0H2+lOhayHWZV/Y4ft4UcijF2LnT3eBwAoWWK uC69b7es/aVjRxsUT72cSSRF3+BKjhz5uF1V2jeLvEHf+1bdd99tt27lf1evU95G2LWDecKLnz6j gcnl+cKWyk7r7isXzp2p3VBkkNXaf/uA0Hr2N/yjS1g5CWNbsXG3ueill7gPnmnIJ0T6uEb2wpGj 7UTdlzfrjXmkVO7rkKFRR+aQPxIyK8vflFkcsXsy87Z02q1XLl048+sjRTvaa012X3/4vnhYr9g/ +SqOuZDvNuqemMNcfIUMddYtWLtHretwPF0cl98soMFOYVa99CNS+kwDBS96UlPywtGP27kn+vLZ +JTK77vvvlt0gAyNJWTBTXKVHxa+rfQTBSG3hvMqXfbdVy6dO/PrDUU76Cek1x70bR0ZoJthLL4c 7v6ayjceq9i7bFVxwap1ta9k9XziIgWxvR9N6/FmbjhoJuaiNYXETE9LSE7GOybXa6ti69B0Hm+Z ktz85QVz2LE3Z9WDMnLcN3AxPEzn8UZPPdrph8xvzaHtl64s03JF56/9qzw3ht5E2nWik/JU2Gau ZXO3Kn0f31gb/WRNF4PV42ETDkTfw94Fe+gcIZ0B4OdMFNVsXthm0tGtjX3/RT9O06kGmoe/biMz XWfXM+mkhcNEP89V9jnGKieUhV0zjNOcCX3HZpPytEPsehydQ2FXP+jESbOHvwBL1HrWHxe7WqXr GWJdV1TTKWuP3URTShv7LAb6qKXzwrRDdE863Uuv13Fa0zA/l91soTO+7GKAWt/Diumjk9qlo1fw XGZanqaZ1uxxsCX2KRNaCT+xQwdCuG5j7+A/H2vobD37+E/UdPqd5qHjQyeFhDmTanYdwsPP56h6 B+ggcq18f9jcCFc76P9BShMdaU0zHR++mECWkWbb6fyRrJZN4Nv7mulxUW2OecKbXiOdvuPN47Lb 2GK39RlkXGWPPR7zqdN5vBnpoV3JrtzQqxr0YFD1xD6rNZ3H2zCbseG09PXosrLjXGtM7KxJas93 s9c9myTlDBbfsJuqafypZMGYhmw2qS0sCkOfQ5jvZqdz/KLWsfk1awd9gfuWSkMfzUNjN/tGh91I oyB/SS2sHL7wgP8cdL67OuxbIAEZonrKBzsWrPnFYWT94d9dhh09lXzkY82qNND3Fzbf7V84dSP7 SorHUu27VEkzaXtcw+ZqmYx95camU7Ge00u4jh6DvxhSaWBBXFiE0Okvj4a9eFwN8wU7Fqz5xU6v r3L85WU6dAbfNVNaVyVtqtAA//hwOhPtkKcjoD90ENmPCXPsx4Tt/LsvDfm0P8EsvproA+3piJBK uJY7unGKz5LgeKPvukaFzP+LylPsx8hu03a80RZYWkdefbJGelUjHgv/5jo9xxs9G9L6DzhFZXMc zhQmBJnpf1fJ/grfky2V+mcWWGhyu50kUxowX8gyZQbnGQ1hwrOxygnNI8q62+n0ZGUHtn6s/rhd wT0Ma5rX6XSFFBOWR4wE3jUr5GcSaB+DhoNlivBTCmOw+JvvdrvpBHnAcPs3JOJxrOMExxv9NYzx jze2KeQVmoiREcoU43gLf4Umpj8zPXYnRg2lQgACEJhegdT+nsn02qF2CEAAAtMlgNg9XfKoFwIQ gMDUBRC7p26HPSEAAQhMlwBi93TJo14IQAACUxdA7J66HfaEAAQgMF0CiN3TJY96IQABCExdALF7 6nbYEwIQgMB0CSB2T5c86oUABCAwdQHE7qnbYU8IQAAC0yWA2D1d8qgXAhCAwNQFELunboc9IQAB CEyXAGL3dMmjXghAAAJTFwj8fb2pl5JuezpuRLhVTlbmF7Jvg63Yx8Vfhv/i8tyYuNZZmZLMv8LQ TIyUkK3DXhf5y8g9M8auIuO2nLE3IDVMAAdxGEmkBNdNz4b9v5w4l2Ltkn/Z+dDEebA17gJDf7b9 E7sXw0TLd1fvKV78yEQ5sC0xAre6Dv/3IL3R3URL9uNnM74wa6Ic2OYXQOz2S4j36G6r+/H7S58q k+fxdbrP1f34LHlIXSYXbjM2eO7ESxfn7tt1l+7ou9+qKKN36XIPnKs51nDhk4Jnn17wVl/B/rLi cW505W6rqbm1ee+mWO7s5R08ffylV9uulf7b8a3LfDc+cw+01fz8d8T/mpLcueC+B2SbVrH2h2yi N3767OaC3ftZs1Nu8Q51vnj0zdVl++XsPp/0tp69dVU/n7Wx7Gn/SNGBey9v+zNFn7zoGxpvb1vD Ud2pexT/tvsu4ytXkm5oflz34VOVT+f5XuXezqbjhnf+VLJ/f3Fu6g3PQEtdnflL+w9smcMfWIPn ml5q/mTr/mdW8XdNI9e7a3586v6y/0UM9b6XgP9I/rsDuwZ+877wUkq5Y3KCBmO+ewKcRG36qLli zwstvhurOq/8+54Kun7lulCd90LDrqqWP3o/6jxYtecT9hHT/X/KNxxsu1H0wD3O7qMVey65xm2X 53flBzvoXRljWIY6XlKUVy0ueuCubP62vnxRnk/fO1hxsPFCdz9durtf2bOrpCj/0OkBujFoE9t8 9YrJFmHaIobmJXTXTAlprKp4/R2rUMvQO/o9VVW7dOd9tyh2X3lhT8XvPyeukaFxm//p0V1Gsrjg ruyPLiXj0FRUHf3Ifzicq9m1dkf5lWWPrUvBwE1HJIsMVB18/tLoy2RHVVX5a50232BdOlVeVUFu /8LIS2DkSJ7tvOR/KSX08BG7cJx3iy1Ob6e7plRFtr1tcZetlBDn++8YWBPa37o6tGbNfOIdoHeF L218+Esrl3k8z2Ty4+N0ENkzP3i+bBkh2z1q/42Bx2i49Fl6O2FhnzG2Tirpo/eNhNMdff7poNxZ 9JRb1nDytZXC8aL7wb6s5W9c7v/XLXkkZFPQbqm2Il3xjIzsOf/+T7cW0I52n+c/4De89f7L24WR ousHHsyT5u71DY2Xvr1x9S8fXUPPYosve/Ym4dDME8ags27nhvKGUnqz5rJVqTYqvvbmFm3kSNXv rg7J2cukj75M6NLWYfZuyqWD1Xv+JL0Z3lfnSzf4XwKBR7LHUx7byyIZzXDePQ2jkvfA1whpMFnY +dyVM8eJqrZWRQ6+2UtXvdZLxwgpWZPv7j/9WNH+7uu9hwqz97ST9j3LC8uaLp/c99i+k74TdpZ7 oO5QTa/vtJCuO//397Y2dNPtzqZ9ykM1NWWFGWwp3NnUOUg3sxz9bfvkfGKGvK6tn6a4e5uUZSe6 zp2gqc89v4ujlZl3ZWSUdY1Ww+9JAi4iZeYXygi7V7lvCdjkT0rNR+mDTyrIsQ7+xHvgTEW7ppHe E7T+wvvMwnKhmd46lMvNFIbmd6amwizOQMyq7Ix9Tb39wUPjHmg5XNM2OjLTNzS3Z5PuE2Vr99DA bUzdwM0Op/nLttL7g7/1B/rU3Wc8Rs9wGjXmqv+0emnC0MUKs2z3+jl+594TZSNH8sXL/Esp9Hhm Rab0gtg9DcOXufCr9IJaaxcNEdffOmhWP/n445tVpOLsECG2y+fpDbPX5Ek8Nyzt5ss3Mhf+3VF6 F3V2d2Htd9cMf3q5vW2QHavC4vroeNUrfwq4dP+J0WDzsO3ODwxV5eW372vt6zFWcw071i48OeAm Q+cUSx5tK9SZesyG6oV7Hl1S1+30eJyG+l1FG3aptbpvbPkOf/tfVWPzdxdl+2vxPRqaT7W0tbS1 tbWcOPwPu9qJbvNX/TmETS2+5VzvaAv9OVLlMf/BEkKqeuhIDF6uIpzicYVSRl6/YKHhwtRsIJqH cuk0ET80txasOaqvpjcG36Fv3rnmrhvBQ+P51FxR/jv/dAXr/TQNzQenqsq4XfWEq325bE2qjMI4 7cx9pJIzN3bSIGwxniKKksdL6GAdY1Nczt5GQpTrl44431VcOnIk3z3Mv5TGKTR1k4XPwKnb/tRs eWaeTEUUb/3+5cfvOkhI88rcXM/XCNnW69zvaT9GVAZ6ccnFXxnMItKV8r9dJyPvlyg3rSnofpf2 d1bwmM0bm8BBiKb56NNyunXva5aPGvKPG678zax/byelxv2Kxdlk6c4fVL7ScPzUu9uVrABth+1A 8Xz6ZJ5SUfHBqm9uYtMAIcvBbfSlMrpcs9kJEa4bkaBN9N7txf6rnKPZU+OZdOmD9E7f7/zhOnf9 DUJ2r5DOufNJxa7jJufurFYDqfzHlXw32EVbyeyCB5WyeeTURuWmVZmk+wxNCxyacD+/gKhDY66o MHMcMZv3/KzzibI1vvHyNyXFHpc9spVUtFncu7p/blA89e/SOXMrCfl/lwe/vviSmShkK+ildd/Z 9ZyC4m3+I9nDv2yyUqyvkZuL8+7IRgnIkVm4WU3qjRfOv8k+htOYmfsAPRM/e/7C+WNEvbkwODp7 2JfJbwacXRNnnTDvkbPWTNrX5vBzIPIa31UcobmfksqSkfOsHOEcOkt6F52rWbtwLl1y5i6vMBOz 6cpnLL9i4wMscNPFc5OGFhJYGZ9M/5OZXMP+xdGqVVSU/GO3b1IgcNPw8KmnfV9P8e+ZSo+SfKWC tHd2XvjPeln1etqR/IeVxNx6/qKxgcgeud+n5OuRlzl5XAH9c3Yp+dHIKSonpHyub2S6AnIQIurQ EG2z5fJlezWdx1+7qzPoEAlqVEqszL//EY4YLrxzvrmdPFG8kJ46PKbl6t84+86FNqJQLg1+uxz/ SE6JvkZuJGJ3ZKNE5Fi4WkYvvDxaUiHTPkI/hhOS97iGqyh5tIJwX19ND8qJF+n2161Wq83aQy9z coYe+sxqfX1n0DnVPDoH856/FDaH8ambzo58TN8qrP4AbDEamisen80yOcaM1v7dhcfASW3pV1cX 0b1u+GJ84KbgnVJvTbpWWdpeXrLtGHlSxj6DS5asVZGGkg27iEx5fxDxWH2TrniZjYy9p1lDP/j0 2ISRWRGUVdShUZQ8Qr/KOWfv6x0yYlj7rRo6G5TCy5xlu+mb0IaSBjp7xX8LdcXG3aR+x6N7DKon 1gaH7hTu5SSbHnyGN8mdkC1mgczcws2EXe5TPny/UNjKkq2kyky4rfRqWHjxnwYnSefnsnPbuffI yLx78nPn+w5b3wdGumUWDacVGw4XmvYU333x5WcqCKktuW/5vO+SPYpnDhdWf7+YXG3OX7tD0dj3 0EpWtC8Is6fjLYZ3znbSORLPrVvXP75Su62CKHTsQypb/Jv4lVu3yJe/sibPt4lPSqn/8tc+Sj+d 0KsODy/l+5C5hH1GOkZP7OilsIiLZH4uey+eu/Bu+v/C+fP9DKPA4g4N//5KD4/5xa93aBdsKH/y yFfbDhRH7EayZpi/ZpOMtLdzmg3sjwsIka5YT+e46CnMw2vy+YQ0+m+MMJFGvZ/Ori7cqJFVVS1c v8IXEObf/5iCVDh2CKfhJIt99853PkufLZ4lzNflkKD57Syag34nivhPOWhOYRqDznwoNJp2RRGN 2nTRNJrLVtKKtlhaq7c8qsjnUxUa/cvbC/iZ2pyR2cAsGlqCqhg12lOydmRFpqru+QmbGxHeLgI3 0TzVJsfeVf6oNbJPijyR5NMTbVKv3rzEp5pZ+HUNOVZVwp+G004EDk1An4LcWKjm2LT4yDJNQzM6 svOLD3RoL2w4uKFuo71s1STehkaankxP7ttIr8+0by0p8jVKsvSJUmJo0NDL+0LKiPPIkTzOeCVT r6bUlgz6AXpKO6bvTh7vf9ec7pq4//fnffHrRYsnzpPIrc6awhzS4Nq7MtPpdGVlSyVB79FemkjC UhPZHpHK/vym8ze9pyau7H8uevDeuQUT50nk1jQdGkrq/bD5L473J7bNWvFMBn5tZmIj/1bEbr/E jHp01mTk/Mlofz7Fv1cwo8bE1xkMzUwc1enoU9D52HQ0AHUmQiBrjV53K+wb2omoCWVGKYChiRIM 2ccRwHn3ODBIhgAEIJDEAviOYBIPDpoGAQhAYBwBxO5xYJAMAQhAIIkFELuTeHDQNAhAAALjCCB2 jwODZAhAAAJJLIDYncSDg6ZBAAIQGEcAsXsnczOaAAAAg0lEQVQcGCRDAAIQSGIBxO4kHhw0DQIQ gMA4Aojd48AgGQIQgEASCyB2J/HgoGkQgAAExhFA7B4HBskQgAAEklgAsTuJBwdNgwAEIDCOAGL3 ODBIhgAEIJDEAojdSTw4aBoEIACBcQQQu8eBQTIEIACBJBZA7E7iwUHTIAABCIwj8P8BjN7UXbcj nZUAAAAASUVORK5CYII= --Boundary_(ID_UQnfOsK19lMNJYbtgzxRtQ) Content-id: Content-type: image/png; x-mac-hide-extension=yes; x-unix-mode=0644; name=3G-upload.png Content-transfer-encoding: base64 Content-disposition: inline; filename=3G-upload.png iVBORw0KGgoAAAANSUhEUgAAAfkAAAGCCAIAAABsMS2xAAAXfmlDQ1BJQ0MgUHJvZmlsZQAAeAGt WWVYVU3Xnr1Pcjh0d3d3SXd3CsLh0M0hRCxEREJBUBABURBBpASUMigBEQMMbEJAKREVBAHl26A+ 7/Pjff99+7rO3jdr7rVmzazZa88aAOC4ToqKCoPpAQiPiKU4mBrwu7l78OPeABxgAYxAEvCRyDFR +nZ2VuB/Xt9HALTd+FRm29b/pP33BgY//xgyAJAd0uzrF0MOR/B1AOBmchQlFgD0tj2hvbFR2/gU gpkpiIMILt/Ggb9x6zb2/Y0f7HCcHAwRzjgAeCKJRAkEgLiAyPnjyYGIHRoiAFjGCL/gCACY+BGs Qw4i+QHA4YNwpMPDI7fxCQSL+/7LTuC/MInk+49NEinwH/x7LIgm0rFRcExUGGnfzh//n7fwsDhk vnYuVuROjIo1cECe7Mi8sQfHmjshmBnB8kFxZs5/sFFikJPrNheRu0X42tgimBHBfuQYQ2QuAWIH ig+NtNy2s8054edvZIxgZFVABTHxjn/xlcQgQ5s/HLcQksV2zGgRTiuJgqDf/fZFxdpt+7Bt83VE mI3VH7wUQDHZto/IYax/jLEjghEfYO5YitO2HPEZlg0INjFHMNIvbBAVtrPmtjkOlDiH7bEII9jP P8L5r26aH8nIEpFzI/JCYAUMgRHgR+6RIAz5UUAw8EOef+Xkf8kdQSL4BCKAP4hBNHYY3sHJlL8Y mAASoh+ItMv80TfYkfiDeETr51/eo4WWhb/4j47vPxomYGrHxh8L8tXys/Kbf9n8dH/9whpjjbBm WBOsxF8J0tPvUVB2/LNERuMP4hBb/kjff/3596ji/mH8W/p7Dhx2tEIRRvDfvoHLjmfB/9iy/Gdm /swFWhStiFZBG6C10TpoDcCPZkVzAhm0MlodrY/WRWshbRr/muc/Wn/8lwEBO3MVv+N9KPiAeI68 1bH+CbFIrIBhZNQ+SnBgUCy/PpIt/KX5zSPIstL8ivIKCmA792xzAPjmsJNTINbH/5H5vwFA0wQA qvL/yIIGAaipBoCz9D8ycXnk/aUDoG6GHEeJ/20Pvf3AAAKgQ1YaB+AFQkAcGb8iUAVaQA8YAwtg C5yAO9gDyCAI8ZcC9oID4AhIBZngFMgDheACuASugFrQCFrALdAF+sAgGALPwRswDqbBPFgC38EG BEE4iAZigjggPkgEkoIUIXVIBzKGrCAHyB3ygQKhCCgOOgAdhTKhXKgQKoWqoAaoDeqCBqBh6BU0 Ac1CX6F1GAUTYWaYBxaF5WB1WB+2hJ1gLzgQjoYT4RQ4Cy6Ay+AauBnuggfh5/A4PA+voACKGsWK EkDJoNRRhihblAcqAEVBHUJloPJRZag6VDuqH/UUNY5aQP1AY9FMaH60DBJLM7QzmoyORh9Cn0AX oq+gm9F30U/RE+gl9C8MDYYbI4XRxJhj3DCBmL2YVEw+pgJzA9OLeY6ZxnzHYrGsWDGsGrJ+3bEh 2P3YE9jz2HpsJ3YYO4ldweFwHDgpnDbOFkfCxeJScedwNbgO3BPcNG4NT43nwyviTfAe+Ah8Mj4f fxV/B/8E/xG/QUVPJUKlSWVL5Ue1jyqbqpyqneox1TTVBoGBIEbQJjgRQghHCAWEOkIv4S3hGzU1 tSC1BrU9dTB1EnUB9TXqe9QT1D+IjERJoiHRkxhHzCJWEjuJr4jfaGhoRGn0aDxoYmmyaKpoemhG adZomWhlac1p/WgP0xbRNtM+oV2ko6ITodOn20OXSJdP10T3mG6BnopelN6QnkR/iL6Ivo3+Bf0K AxODAoMtQzjDCYarDAMMM4w4RlFGY0Y/xhTGS4w9jJNMKCYhJkMmMtNRpnKmXqZpZiyzGLM5cwhz JnMt8yPmJRZGFmUWF5YEliKW2yzjrChWUVZz1jDWbNZG1hHWdTYeNn02f7Z0tjq2J2yr7Fzseuz+ 7Bns9ezP2dc5+DmMOUI5cjhaON5xojklOe0593KWcPZyLnAxc2lxkbkyuBq5XnPD3JLcDtz7uS9x P+Be4eHlMeWJ4jnH08OzwMvKq8cbwnuG9w7vLB8Tnw5fMN8Zvg6+OX4Wfn3+MP4C/rv8SwLcAmYC cQKlAo8ENgTFBJ0FkwXrBd8JEYTUhQKEzgh1Cy0J8wlbCx8QrhZ+LUIloi4SJHJWpF9kVVRM1FX0 uGiL6IwYu5i5WKJYtdhbcRpxXfFo8TLxZxJYCXWJUInzEkOSsKSKZJBkkeRjKVhKVSpY6rzUsDRG WkM6QrpM+oUMUUZfJl6mWmZCllXWSjZZtkV2UU5YzkMuR65f7pe8inyYfLn8GwVGBQuFZIV2ha+K kopkxSLFZ0o0SiZKh5Valb4oSyn7K5cov1RhUrFWOa7SrfJTVU2VolqnOqsmrOajVqz2Qp1Z3U79 hPo9DYyGgcZhjVsaPzRVNWM1GzU/a8lohWpd1ZrZJbbLf1f5rkltQW2Sdqn2uA6/jo/ORZ1xXQFd km6Z7ns9IT0/vQq9j/oS+iH6NfqLBvIGFIMbBquGmoYHDTuNUEamRhlGj4wZjZ2NC41HTQRNAk2q TZZMVUz3m3aaYcwszXLMXpjzmJPNq8yXLNQsDlrctSRaOloWWr63krSiWLVbw9YW1qet39qI2ETY tNgCW3Pb07bv7MTsou1u2mPt7eyL7D84KDgccOh3ZHL0drzq+N3JwCnb6Y2zuHOcc7cLnYunS5XL qquRa67ruJuc20G3QXdO92D3Vg+ch4tHhcfKbuPdebunPVU8Uz1HvMS8ErwG9nDuCdtz25vOm+Td 5IPxcfW56rNJsiWVkVZ8zX2LfZfIhuSz5Hk/Pb8zfrP+2v65/h8DtANyA2YCtQNPB84G6QblBy0E GwYXBn8JMQu5ELIaahtaGboV5hpWH44P9wlvi2CMCI24G8kbmRA5HCUVlRo1Hq0ZnRe9RLGkVMRA MV4xrbHMyCbvQZx43LG4iXid+KL4tb0ue5sSGBIiEh7sk9yXvu9jokni5f3o/eT93QcEDhw5MHFQ /2DpIeiQ76Huw0KHUw5PJ5kmXTlCOBJ65GGyfHJu8vJR16PtKTwpSSmTx0yPVafSplJSXxzXOn4h DZ0WnPYoXSn9XPqvDL+M+5nymfmZmyfIJ+6fVDhZcHIrKyDrUbZqdskp7KmIUyM5ujlXchlyE3Mn T1ufbj7DfybjzHKed95AvnL+hbOEs3FnxwusClrPCZ87dW6zMKjweZFBUX0xd3F68ep5v/NPSvRK 6i7wXMi8sH4x+OLLUtPS5jLRsvxL2Evxlz6Uu5T3X1a/XFXBWZFZ8bMyonL8isOVu1VqVVVXua9m V8PVcdWzNZ41Q7VGta11MnWl9az1mdfAtbhrcw0+DSONlo3dTepNdddFrhffYLqR0Qw172teaglq GW91bx1us2jrbtdqv3FT9mblLYFbRbdZbmffIdxJubPVkdix0hnVudAV2DXZ7d39pset59ld+7uP ei177/WZ9PX06/d33NO+d2tAc6Dtvvr9lkHVweYHKg9uPFR5eOOR6qPmx2qPW4c0htqHdw3feaL7 pOup0dO+Z+bPBp/bPB8ecR55+cLzxfhLv5czr8JefXkd/3rjTdJbzNuMd/Tv8ke5R8vGJMbqx1XH b08YTTx47/j+zSR5cn4qZmpzOuUDzYf8j3wfq2YUZ27NmswOze2em56Pmt9YSP3E8Kl4UXzx+me9 zw+W3Jamv1C+bH098Y3jW+Wy8nL3it3K6Pfw7xurGWsca1d+qP/oX3dd/7ixdxO3WfBT4mf7L8tf b7fCt7aiSBTSzl4AhdzhgAAAvlYitYA7UgMMAUCg/V0b7DAAQEEIB8FoZC8qDizBftAKwZAj1ABz wzkoFlQD2gcjiFnGjuDu40eoVqn5iCE03XTC9IWMQkw1LLqsy+xFnHbcNDwP+TIEnIW4hN+IXhL3 kxSRGpVJk+OTb1A0VRpTiVWjVr+oqaLVqW2iM6TnqD9kaGnUZSJvWmSOsgiw7LFms4m0bbVHO1g5 Zjs9dmF0dXBLc+/0WPEU8LLaE+ed73OL9I4M+Qn46wS4BoYHpQYXh9SHdoU9Dh+LWIxci4YpxBiO WJE4xXidveYJ7vuCE0P3mxzgOvD5YN+hssMHkjyOqCdzHgVHZ1KGj3WmNhyvSCtJL8w4m1l4ouhk WVZtdtup/pznudOnv51ZyZvLf332YUHXuRuFjUXXimvP15U0XLhx8VZpd9nApeHykcujFR8qZ698 qvp6daV6ueZr7ee6b9foGuQbbZpCrx+9UdLc3DLY+qZtvn3jFtVtljv8HRKdil063RY9bncDeuP7 0vtL7l0fGLzfMVj2IOmh5yP1x0yPF4d6hwufRD81fsb6bPb5zZGsF74vlV9hXr14XYOsJ+d30qPo 0edj+8f5xjsn/N5Tvb82uXsKN9U4TfpA9+HWx5AZ9pn+2dg5/rn783sXhBYefIpb5Fns/hy6xLzU +sUTWT3l3yy+fVnOXVFYefx9z/fZVdfV7jXONZ+1irWFHzHrXBt1v4S2tnbizwikgQOyC7wHsUBk 6A4sBpeixFAtSPz5MXPYu7jr+Faqh4Q5Ij9NEG0PvRDDSSY8cwYrD9tNDi8uFHc1rys/rUCX0AER CdFR8VxJJalqGTrZULkBBRnFbKUVFXfVDnVBjcOaY7t0tAt1vuvZ61cYrBuZGxeYvDcTM4+0qLac tRay2W2bY9dvv+Eo7uTknORyybXP7YMHaregp5aX+54Y7zSfYlKdbxf5qd+k/5eArSDaYO4Q8VDl MP1whwj/yNio9OhCSmPMYOxwXHN87t7QBLN94on4xLn9Dw80HCw6lHI4Imn3EYtkzaPSKbzHGFNx qZvHv6bNp09lvMt8fuLhyZ6s9uyrp4pysnKPnI47E5EXmO9z1rPA7ZxroUuRc7HzeecSpwvOF51L ncocLzmXO162rbCuNL9iVxV/taL6Uc16HV+92bXIhjONTU1Pr680s7QotTq0RbVn3Wy8NXJ7o0O4 06brQHdtz3gvsY+rn/ke7t76wOf704NvHww97H10+3HTUOVw0ZPspynPEp4Hj7i+MHqp9IrvNeH1 8puxtz3vcke9xsTHVsZ7Jk6995yUmPw+1TWd9sHmI9PHFzOFs+5zbHMj86cWLD9hPt1ejP4s8nl0 KfuL3pclJPpuy4Tl5hXf73Tfb646r35cS0fyxtb6p03qn4G/2reUtvZu3dyJPwyoASdQAHZINVAJ xpE9PQmqhSHYB76H0kK1IRXSS8whrAJ2DteKP0EVRvCkdiDa0jjT+tLtpT/N0MT4hGmFhYVVi82D /SBHMWc711PueZ4NPhw/m4CEoKaQtbC3SIRooliq+CmJc5IXpSqkq2SqZK/IlctfVMhTTFVKVA5W cVU1UpNVZ9EAGh8172td21Wkna6zTzdIz0Ff10DckMXwp9G08YBJrelps0RzLwtdSyErrNWs9aBN LbJm4u29HUwc5ZzYnFHOSy7Drpfd4tz1Peg9xnc3eB71ctojsueH9wOfElKMrymZi7zk1+tfEBAe qBfEGjQf3BWSFxoctiucGD4ccSJSN/JrVHm0IwVDaYkJjGWLvRe3L14qfnRvToJJwua+G4lR+6X2 zx6oPhh5SPHQj8PdSSePeCXLJP88+jCl6Fh4qs5xuuPjadfSUzJcMyUzf54YPlmetT/b4ZRkDirn bW7b6bwzlDzbfKWz3AX4gm/nJgqHirqLG8+XlxRdOH0xszS1LOVSannm5VMVeZXFVyqrmq52VD+s Ga/9Uo+5xtkg12je5IdkmbLmnpapNmK7wk33W8m3a++87sR0yXXv7sm429T7rp9wT3GAdL9mcOOh 4aPDj9uHPj0Rfur/rPz51xeuLwdeW71Zftcx1jrxYcr+o9Bc2GLIt/h19+34/z4j2v4mYFUByENq T5fHADhUAZCDnIOIUwBgIwBgRwOAkwaA34oD+MoKgGKU/vl+sCMVpjVSDx8C+aAOqRjfgWWICAlC mpAdFAQdhs5CdVAfNAatwfSwOGwAe8Lx8Cm4Gu6D38NbKC6kjnNGxaJOo5qQuu0bmgWthvZAH0Rf Qg+gvyKVmSkmBnMB8wDzAyuOdcUex7ZiZ3HcODuk0rqNW8ZL4wPwZfhRKm6q3VTFVGMEIUIw4Rrh B7UhdS71BFGeeIz4mkaeJp1milaf9iLtLzofuj56GfozDIAhimGC0YnxIZMRUwfzLuY2Fi2WDlZT 1idsXmzzyDpm4KjiNOQc5TrILcDdyxPGy8h7iy8QqSB6BWIFRQTfC/UJN4qUimaI7RMPlHCVNJKS l+aXoZb5Ljshd1++SaFY8ahSuLKzipaqoBpBbVl9VGNQ845W/6432ou6kB6HvoSBhqG1kbcxxeSY aYFZnXmvxTvLDWsOGxXbPXbH7K85vHZicDZzOeZ61+2D+9Zufk99r6A9ed6dPku+omQ/v3L/6UCp oPjg3lDWsMjw3kjOqNjowRjR2KS4F3vlE9L3Te6nHBQ69D7pcnJEilYqy/GV9OeZfSdbs+tzKk6X 5lWerT3XUNR6vudCT+lYeW7l7qvsNU/rkxqVr4+1nGt3us3cMdZ9q/fUvUODUY/2D8c9y3lx8/W7 UdSE+VTlTOqn/JW4H8rrLzY+bb76WfGrdCd/sCK5w3In/mfBNXAPOSlYgxggCUgfOQuIhjKgS9Bt aARaggmwMKwDu8OxSPRr4AH4AwqNEkTpobxRh1ElqE7UezQaLYa2QEei89Gd6M8YPow9JgXTgpnH 8mNdkKr7LnYTp4qLxtXhFvCS+BB8NX6RSoEqnuo2AUOwJRQRZqhVqdOo3xBlianEMRoNmnyaFVpn 2lY6TrojdPP0LvTdDMoMlYzcjHlM9EyZzETmTBZaljxWXtZqNg22e+y72Zc40jmFOO9weXL94i7l MeaZ5c3iU+Mb5c8UUBYYE8wU2iOsKyIuSiu6LDYq3i9xXbJYKk06XoYsay+nJy+nIKjIrESlDCmv qnxWnVObUZ/VWNRc3oXRZteR0tXTc9MPNThiWGjUYPzAZNoMMudDshjJKs26xmbEDm2v5ODjeNbp iQu1q7VbgnuKx+nd1Z53vN7u2fRhJWn6+pCz/G77zwVyBzkEZ4R0h0HhOhF7I5uilijiSJ4qjXu1 lzHBct+xxLEDvgdXD2cfEUvuSHE79uN4UbpWxsSJrKxd2Z9yLp52zePIny6oKzxU7FyieJG29Fe5 eIX3ldyrvbVQvUHD0aahZrnW3Jvw7X2duO4zvar9o/dzH3oP8T7ZfP7h5d03taNtE31TYzPZ868W i5c+fn22HLOytnp9J/4iyKlSArgI+sECxAipQp5QEnQZGoS+wOywLnKakw23wVMoepQOKgx1HvUI DSNveARyBjOJ7C7JmMuYGawMNgZ7E4fBOeBKcV/xxvgi/Dcqa6pqAjUhgvCMehd1DZGLmE2DpUmi +UWbRAfTZdCz0ZcxyDC0MZoyvmWKZqZmvsxizDLDmsOmxfaB/SyHJSeEfIfiudW413hu8h7kM+Kn 5h8RuCAYKmQjrC2iLCopJijOI8EtySPFKy0mIyurKWcu764QpZimVKHcozKlRlBX1iBrFmmtagfr jOv56U8bUowhk3wzafM+S2+rdZtsO3H7W46WTuMuka6/3I/tZvS8uEfde5BE9t3yKw3QC5wMzghV CnsfkRmlET0dUxBnHL+ScCXR6wD6YPlhi6TF5NwU5WNvjx9NF87oOxGaRZNdm2OWO3HmUD7v2Zvn nAu/FGeVSF3oLiWVbZTnVahWPqkKrMbUnKtTqh9sCG5cu57XLN1yt82rfflW5h3xjoGuwB7c3fI+ 2/4fAyWDVg/hR+eHuIZLnko+6xgJfIl/1fYm9B3taN2468TCZPjU1AfHj22zzHN280cXaj/1LQ5/ frbU9+XG11Pfwpall+dXSr6bff+6mrbGvVb/Q+7H1XXh9ZINeCNg4/6m7ObxzeGfXD/9f9b/XP9l /attS3Qrdzv+MQFKittfDwARDZDjx9GtrW+iAOByAfiZs7W1Uba19fMSUmy8BaAz7Pf/HbbJWOT8 uzhlG/236/8AleWgQatZQGAAAAAJcEhZcwAACxMAAAsTAQCanBgAACAASURBVHgB7Z0PWFvXffcP HcQiiYgd1yyz44GDnRq/8aWFZOA0diacZebJamkttH1tnDc0q/CTZQZniRneQjfoyuRmsWGZHyBz 5DdB7jJ5m+XGhWUGMsgf8WTCscgbSAI11BVLRCwSqbVkSxvvOfdKgP4A94KEdPH3PAm699zf+Z3f +fyuvvfcc2UpaXJykqCAAAiAAAgsawJfWtajw+BAAARAAAQYAWg9zgMQAAEQWP4EoPXLP8cYIQiA AAhA63EOgAAIgMDyJwCtX/45xghBAARAAFqPcwAEQAAElj8BaP3yzzFGCAIgAALQepwDIAACILD8 CSQv/yHecCP0Dff87Nw7g+S29Hvuf7hw69opAMM9p0+0mswf29LS1t2vKSsrLVwTnH/fWO/zJ16/ tiLr0af2ZPCHPKOdjS+/RdK/fqC8UDHlKHhjtPNk8/nLDz3+TGHWbCbBDSLtsY6az6c/9PhjhVmR jkeuGzzbbLxoX//1Rx8rzKAWw+3Np961p399TznvhAb28luX0+/bU74rzKdv9OTzL18m0yMN78A3 3nu86XXPihWBQ4r0OzMLfu/hzWsWPsyAq9lffWOnnj8xTAM7sCdjUf2MNWqKKsjjoy/mnwkaBSGK 9AfV38rPUM4ehOgjnrHO195OuWfn9s2pZw9/V31O2X1evz3krBLtDIaxJUD/3SzKMiLg7dapZp4x 6oZuLxuet6M2qJ63qR7ij00N32lp4OtVZqe/zl+janBMGYVtWBrUtFWDZQ6TsDZhFUJHqgZL2JG5 Kqz6UhawWs/Ha9dxQvhN/K6zicVFtIahCC6cFv7g9EjDbQI0eJ/Tf9Td9nDb6NU4zcIgugMpWJjr kbZqGrK6wTLLKIhhYHEdsLC83XzqOZ2Z7ti7dbRHFb+9sJjRKqYEsIYz/SZeDlsT7/ygqosORNdm MRvZu91U8bcWF3H1n9hZw+o5bZN1ZMRsrOUHW/+XrYP8RuBPijCHTUsJVBB/zYqQGwCPzzfDZFEz RB/vKvXmNOqQ/S+lbNpexMxN79hoOBOD7Va+cdebH7sI8Yy8aaK73CNfZ1P+0JJC+KDX3ZxKjwgh hJoExs41dVisVoupqZK3MF34+USYafQqlHlv2EZGbPZtQVB9M4ELnflo3azduroM9fTgtx/eEjwK q7lDz1/kiM743qyt5z/g6z/buK8wZUcFQ7xBwc6XNV99iHruqjKNzt8cFnEgAK2PA/TYdem6dIEp utpwaFdufvHBWmGKSHxdLx1nnXK1bc3lWzMy8oufNevYW9760cjsesFazCie9rpyjWZf+eGDmqSU 1JSUwvLGQSqpwcUz1nukvDCJL5ryxr5xv/vBzpP7CnOE+sJ9dT2jHtZuov8IFYyUlKScfc8b3wn2 NHPP09lcd+RUb3ioiox8XoBb3rcRjzB21q71XTosW38r2y6+NyPZNdpzeJ8QVc7BxvYZUt165Jny nCQWwsHmHj4m1ia4cA8W5m7dmru7vEbgSc08g6c1hYWHm8+eqtMkJWl6XSR84IKN5uBJ1p2r/7BG U6g53M/6GG8up63L20d9oz3TWPYdPsmoeEZf/Ku/qKjSD3n8wA8fqduXw4DnaOr6J3gGfm6pKUn7 jjTW7dMUHj41GBSzZ6SDDb4yf+PUMhB3/w46iq35hY/9DX8zZP1o2OUZrNPQlB4ZZlGRwdN1hTma U4O0+vS+Qs3BusY6yoYVzcnesSD/xNWmrmhlp9qMotykKaW79V2DvLsZR7CZEARietcA50tOwOt2 Or2T3qEOQ3Upv2hTqndM+lczqttsfDxuh91O/7PRYg+6kXdam/iTUm0JVPtrVHRVxNkUWARSV1aq hauIqoGuZ1ib2FucreE4zOwCQq8ppVMWpVb3ZGAZgdPpDbpK3gtzGFhyodZaoR1bc4hEjO+aawgE NdPEa6pkoVS3DVmEJRs+gFLDwJCJXQW46g6vrU0ItrSyWqtim6zSbWFB86U00LvWMDDTdYAG0Vbr aKGNefNaCnHqEF+jtvwi0sAnhC64Nsek2784RhooDjubC1MhHrJ38x45ncHY5Meid/sXl9Rm5wzg 2lJhCPxqib1BCIRTV/pDCuXmHjCwHhjkqVBVpgEbS/mQuZaHrW6y0sQIm8KSnZUHSPM4PTpVqXAS 0ZWwgeDlPvvQwMDQgLGahTKVNeFM0BojLZrNJIvteBDAvJ69KZZRSVYolcnEfe7g3nph3uW65vYR YWkmYx1bF3D1Nd6enk7/W0fLQ/oJ0YNfkcZMS5ssZ44ePfMmL2RdJyxj/EyTd9L/Ty9QGeO0xt5X mAU/fWx94bXhlDtUbSajqfvVA9/9xkP3FQgdegb/vYotuWit1Lr5jLlhSnuF4/xfj2t8fHxiwn6N 7X1hm5igu67pDmllco6qkL6Y/+3cuTbaudrUbaDyY/2P1893XaT1xUVbB37aRPtRVRt/ePiZw88d pKJprW+0Bu5IDAPuV5rPCBeGln94PSKNlvoqWupbhHls19vTtzMqg5Ve7M6k/CzSwM/fUsKuQ9bu C6MfvPsuDYaWdy0fj1rMdENV+811V4YFj59fv/X+fX9lNLaZG1TJ/sUlQpdFBOD0CnSm+ZU3zA20 lbX9vYnx/jOsmdby5pmjzZ3d/P0Z3Z9ZvF42PK4ggy1Q+UuXOnsdS/nGghrKiVT+bflW2oewUBRY sgtaNqJL747OV14xmfkry6dfuAOe+Nc1WZs3Z22+5+51QbX8zqdjX4RXoibuBKD1cU9BNAOYGO7r 6ekZHCMHLk667RYmNab9rZbPrjlZL29aRujf1PUP6Zv0DcJEckPwQryXmRGqAIF3v+8qL7O8ygvH Hrh/E2+ypYiJs9V6OaCawmH6987f5BcOlAU7mcXHY18kp5KPzrWqd2Snpqbl7WXryNML8+ptm3jr HNUDrD649DU+kE4vS7dvrKDqZq3Jvp1eodKPWYIEed29bK7edayCSRh3//btv0svJtaWiv3HaBvV 796j/LDvEvXaVV+SmX57Zl6ZsKTvHygp3ZrJus/a/g2maGnBNPzBqM3+p87ONjaN7So52vXfwiHt E9/eusZvRV9CB/7r7d/cS6vPvfry6bZWQufIHGk985OX//UcrfzOI19VbC4166s5eukpK+LydpSU FJn6r0x7C2zlbmV6mnwzf71OI57L7zOpV2/bwivzfUX8E4uAsfCawj9l2ZCxbsZTFk5nbOtoazM2 sac4hBz7556QZRl6bRFa+/+qdj24im4q79rFi33gjAiy8fJnx1TV53aB7lQFNhKIALQ+gZKx+FB+ ee7pHTt2ZP/Vv9G5r2LNps2rmUtFysqvatjNemvZs+3DruQ1uY+VP5a/gSkI4a8BbIMvyg1b+Pd1 6/m3R/kKX+/rZ9iG8xp1KJS+fhu/YXufXw6/88sz5o6E1333r3kD3/sX2Tt/XVqK9ZWKihaTWtdm 9046hNUM1u91Zma6yB6rEjLS38degsum7+jb2to6OoR1mlJjBxWrjpK7g6afyWtzHgm0UhVvW0XW 7uBXdVidSnPPKsX6/7WBbqobzHRpy+22d5uMBuOB9f6orZcczHD8wltMQGcMk9UGSopfMgNy9/GI cH1T534lIKaRB77qPnZBtLbU1JuI9o8OPflEKTHV17RQLJUPcUrf+MDE7TteGLA5HSPdhlraW/3e p+mD9JByzTvFnuXr1g1fYzkyvfzaMDUd/8cX+CcxwW28vAab3uSfDvgPcUWaXYW7dhWX/6ib/0RW +4VPiFdIGOEv7b5+fo1/ylMa/8SVfUJzxpV+6mjEjZXp9LIbck5FNERlHAgEztU4dI0uo08gu/gQ oXPglpK8q9oCl5lfdeDuvkOZ/8TfVFaYjhFT0cY0lVpNTCYmbbSEvI1X3fMdNekykaqdmQa1evUl UxeTa1JZ+Qib4vGlZW/2zZ81pb57nJ+fV389S+E87z+U/Y393H6Ttb5oX2rDfe7OimO0sUqr3kpe Yd04/2t08J2zbT+oYNZpRLFZRWeY9eTYxuLUpkL3/ooWv5cZL8qM3F0ZdN/zCyrXq3f+QWHEz/hn PFStqq9nA1L97mb6l/v9Rwjrmqg1D9Cwt6g0VBpNFdU15NE7PjlTQXWX09n/4CFqQHVYvW6fTrfe UMVGI9jz9TP/mPIeyGEaZrXyMKhqP7KS8GO+5r89mHXgCh9dxmnlg9mWm51z9T56waWeuNpvZCUT zydvF6n306ibTH+RcV1YIsmhl05B7anr4ElzIKRVX320lHS1dpVsDElewIBJNz9Ft7n88bEjrqu0 B/4quTKdv8x/8WuSupZdBonp6e8f3rX+gyoW2qKKcO1QfW39orygcYwIxOMhAfqMIQFb99QzVCoq pUa2oMwX50BTJZvdC0Vdraulc0618FH0GfE4h2aaMV0yWvjHck7/h/VUU05KTUPscanwRK6J/3y9 3WKYOsz3zpY/vLYO2pVQ2JWGlcoB96R7qG2qnmNqSsOJ/GyWdV0qfIh+RqiBTeGT3bR1t7DYYmtj M19CmgKPmIfaGnj3fK262kzvL4Rns5yq1P+UmV6U9AFSfr/uAT3fYPoPp1LrDGZ34IHnzGgjDpw6 CsSmpQ9lA89CaWBCoF6LIfBRKdoJpzbQereVZ1JqcToN/JYwCqeVD0b4lwRem5E+KtaWVtbqTQa2 JjP1dNQfuneI/3gSfcA7GRgFdeg/6LTwT+A5HX3IPGRi9xN8UQkkaGzCs9nA6IS8Tzf3e+FfhH/f UEof87Ji5/9pB2eyBT/G5Y/hT9wJkLhHgABiQMDrdDicTqouocXLPoNjdzgiHJppSttTM1rc029b v9Y3UZFm7h3TR2a2ZNv8YUdAWvxH+ZBmuAs08lJXDvrRoZgXN99RhIGzwUQngIgDn3dgfKtIyYrc UrgYcGqjlV4wHAYtu4rRzx2FGJt53a0V8+++3AxAhMyEeJx312ZioaiahOvYvOYwWGIC0PolBi7f 7pzC3FgXeFIp35HIPXKLXitMxf1/udqh8EuYrY0d5eb6B8/R5dDN36LoxFxdotsxvIkjkETNgs4b 7IDALATGh/svf0Hu2JS9ln6qEyWuBCZG+y9cvOS4Tm5ff/d9+ZuDnlYHAhsb7PvEe9umrVkRjwas ovbKTo+rN2/ZuogvRYpaLHAUgQC0PgIUVIEACIDAMiOAz1wus4RiOCAAAiAQgQC0PgIUVIEACIDA MiMArV9mCcVwQAAEQCACAWh9BCioAgEQAIFlRgBav8wSiuGAAAiAQAQC0PoIUFAFAiAAAsuMALR+ mSUUwwEBEACBCAREar2v71Qd/xM1hXWn+3y8n+H2RuGXfg4HfjAovMY13C78so3m8Cnhe87DayIE hSoQAAEQAIGoEhD1DyDHe+rz9p42mIfySe/GgrwV3fanfvs/NhZVNHQM6Fe+vzuvIHW9vTqs5tlt zqc3Fn3a0DGiXtmcmVe0cv3Fp9aG1hzaHtXhwBkIgAAIgEAEAqL+3WzfkZw8op88lEsdtJcnVWWZ W2+r5voOTTbvojXs6LUWa3pojVPzXho3aJs8upb9FtKRtDzyC+ttvx1c45g8NPVluRGiQxUIgAAI gEA0CIia1+c+1etNZj/f4xttL2ohTQNbvK8TLitdCGD9tmLyZO/nj4fW/Ip+bTjZIHwXh3L9vYT8 cOjXtCqo5iPXoXwR39ZhNBqjMVj4AAEQAIGEI1BSUrIEMYnSepKsoL+s0Heqlv6AnFZvKd+s7H+d WOmv2vBFccsKYvXQhf+QGh+hlwePsLhPFLeoyBXePLxm/mEuDYv544AFCIAACMiTgMhns+PNmtS8 ve7uEWfzY2wlh8n85/xvyNFfj3uznVTfy37TLbhmJTPy/3adZ6S3ixRn3hRac5eIST1tgwICIAAC ILAYAqLW6wdParLL8oac1RkpbrebpCiVXmtjWt6Zbtvr25VDB9Oyr5pGnrvTFFLTrLpSmJa3q9t2 aLvy9MG0kqsm73N3PhxS07xb3J3FYsaItiAAAiBwwxMQ8TX3gR+fC7BqYL+g5u7Q+X9sjtMa+B9v C6+ZHOnQBRpVCj+7Fl4jIgCYgAAIgAAILIqAqHl9QK9DX30eF5vlK9hjW6GE1xC+SqmctolQE2iO VxAAARAAgVgQWJTWxyIg+AQBEAABEIg6AZHPZqPeLxyCAAiAAAgsHQFo/dKxRk8gAAIgEC8C0Pp4 kUe/IAACILB0BKD1S8caPYEACIBAvAhA6+NFHv2CAAiAwNIRgNYvHWv0BAIgAALxIgCtjxd59AsC IAACS0cAWr90rNETCIAACMSLAL6NJl7kl3+/H2Z/TeogvzJwQWoT2IMACIghgHm9GEqwAQEQAAF5 E4DWyzt/iB4EQAAExBCA1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/ iB4EQAAExBCA1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4EQAAE xBCA1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4EQAAExBCA1ouh BBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4EQAAExBCA1ouhBBsQAAEQ kDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4EQAAExBCA1ouhBBsQAAEQkDcBaL28 84foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4EQAAExBCA1ouhBBsQAAEQkDcBaL2884foQQAE QEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4EQAAExBCA1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9 GEqwAQEQAAF5E4DWyzt/iB4EQAAExBCA1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQ AAF5E4DWyzt/iB4EQAAExBCA1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DW yzt/iB4EQAAExBCA1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4E QAAExBCA1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4EQAAExBCA 1ouhBBsQAAEQkDcBaL2884foQQAEQEAMAWi9GEqwAQEQAAF5E4DWyzt/iB4EQAAExBCA1ouhBBsQ AAEQkDcBKVrv6d+XtK/PIwzYc7ZuXyErmpycwlODLlo73N5YmMTK4VO9Pt7KNdxensNqNIdPjfFV 4TXy5ofoQQAEQEAOBJJFBekbb2/9ieFoRStRP+klREGIb8RY0/qoeWD7bd6rV8nqdam+0dMbiyoa Ogb0K9/fnVeQut7+7Dbn0xuLPm3oGFGvbM7MK1q5/uJTa0NrDm0XFQCMQAAEQAAEFkEgaXJycv7m VMhPdF1fcXlvmcXsPJOvJGTsbM4686v2J1OvpqzLWEOvGP3NhVzfocnmXdRb35GcvGstTs17adyg bfLoWkJcfUfS8sgvrLf9dnCNY/LQqvm7h4UsCXyY/TWpcX9l4ILUJrAHARAQQ0DcvD45o7j8MeLr /4eyLsHpxOVBK6nPTj9HiJUQrdnx9ynXCJeVLhxdv62YPNn7Kw3d20CvC7Qo199LyA+Hfk2rgmo+ ch1iV475itFojGjC1fwoYv1sldbaw7MdQn3UCXDSPc6WaOme0AIEZEOgpKRkCWIVp/VCID66fBMo t21r6x56aHtWMhlv1qRrXylvZarvN1DcsoJYPT621kP/8kVxi4pc4bfCawI+Z3+djcWHErV+Nj+z 94wjCycgNTu0JyRo4bjREgTmJCDl2ewMR8lpa7PvoUJPy5oHi1RWj5fJ/OfXBZORN9tJ9b0rWdU1 Qes9I71dpDjzptCau0RM6gWf+AsCIAACILBgAgvU+o9bN2befmTUQ3wT/c37u0rv/PKmBzSk/gc9 9NM2rsGXKrq0BVnKTQ+oSI2+Z4wu17/2UgXR5qwPq8Fi/YIzh4YgAAIgIJ6AlDUc5jUthfede2Co oX9jZmoV3eMqDW/syVISbYeuc8c6dpzTGs7vziAkQ9+hy9yxjhmRSotjd7KShNawQyggAAIgAAKx JSDuczizxODzeNwkWamYvmD4PC43SVEq6Ep9oPBVSuWcNQFbqa9SP+mBj3lIJbwYe6nZoX0hQYsB jrYgMAeBaZmew2i2Q8kKRch6e7JCGVJDwqvCa2brAPUgAAIgAALRILDA9fpodA0fIAACIAACS0Rg UfP6JYoR3YAACMSAABbZYgA1cV1iXp+4uUFkIAACIBAtAtD6aJGEHxAAARBIXALQ+sTNDSIDARAA gWgRgNZHiyT8gAAIgEDiEoDWJ25uEBkIgAAIRIsAtD5aJOEHBEAABBKXALQ+cXODyEAABEAgWgSg 9dEiCT8gAAIgkLgEoPWJmxtEBgIgAALRIgCtjxZJ+AEBEACBxCUArU/c3CAyEAABEIgWAWh9tEjC DwiAAAgkLgFofeLmBpGBAAiAQLQIQOujRRJ+QAAEQCBxCUDrEzc3iAwEQAAEokUAWh8tkvADAiAA AolLAFqfuLlBZCAAAiAQLQLQ+miRhB8QAAEQSFwC0PrEzQ0iAwEQAIFoEYDWR4sk/IAACIBA4hKA 1idubhAZCIAACESLALQ+WiThBwRAAAQSlwC0PnFzg8hAAARAIFoEoPXRIgk/IAACIJC4BKD1iZsb RAYCIAAC0SIArY8WSfgBARAAgcQlAK1P3NwgMhAAARCIFgFofbRIwg8IgAAIJC4BaH3i5gaRgQAI gEC0CEDro0USfkAABEAgcQlA6xM3N4gMBEAABKJFAFofLZLwAwIgAAKJSwBan7i5QWQgAAIgEC0C 0PpokYQfEAABEEhcAtD6xM0NIgMBEACBaBGA1keLJPyAAAiAQOISgNYnbm4QGQiAAAhEiwC0Plok 4QcEQAAEEpcAtD5xc4PIQAAEQCBaBKD10SIJPyAAAiCQuASg9YmbG0QGAiAAAtEiAK2PFkn4AQEQ AIHEJQCtT9zcIDIQAAEQiBYBaH20SMIPCIAACCQuAWh94uYGkYEACIBAtAhA66NFEn5AAARAIHEJ QOsTNzeIDARAAASiRQBaHy2S8AMCIAACiUsAWp+4uUFkIAACIBAtAtD6aJGEHxAAARBIXALJiRsa IgMBKQSO/V2jFHNS+ScHJNnDGARkTQDzelmnD8GDAAiAgCgC0HpRmGAEAiAAArImAK2XdfoQPAiA AAiIIgCtF4UJRiAAAiAgawLQelmnD8GDAAiAgCgC+ByOKEwwWkoCP971vKTunml/SpI9jEHgBiSA ef0NmHQMGQRA4IYjAK2/4VKOAYMACNyABKD1N2DSMWQQAIEbjgC0/oZLOQYMAiBwAxKA1t+ASceQ QQAEbjgCUrTe078vaV+fx89ouL2xMImVw6d6fXxdeI1ruL08h9loDp8a443Ca2445BgwCIAACCw5 AXFa7xtvP9m4L59rJS6vl8XoGz29sahC0zEwYjGe21tQ3zMeXkN8w09vLPr08Y6REcuW+r1Fz/dE qFnyAaNDEAABELgBCYj8fP3VX11LKzpY21pmERgNtB8n2rYDhZsJ2azXcXlv/FyTHlpTufK9FlJp O1C4lpBqi64+753LRbeF1Ewc2r7qBqSOIYMACIDA0hIQN69Pziguf2xPqUZFnEJ43muEy0oXttdv Kyanez8Pq/kVuUbIBiVvpFx/LyHtQ78OrfnItbTDRW8gAAIgcEMSEDmv59n4+OUbfjOFECvx7ypu WUGsHnrRCKnxEQUhHmEpnyhuUZErfNPwmvnBG43GiEZcxNrZK2fzM3sLHFk4AanZoT0tLEFL2Wrh OBKv5ZIlKPGGnlgRlZSULEFAUrR+RjhM5j+/LlSMvNlOqv8ilbSH1Kz0vk/INUHrPSO9XaT4xE20 RVDNXcK0f4bniJuzsfiw5kcR7WernM3PbPaoXwwBqdmhfQkJ+vEJad+RILSS+lslOBkWnKDFnBVo Gy8C4tZwwqLb9ICG1P+gh362xjX4UkWXtiArvEa56QEVqdH3jBHieu2lCqLNWR9Wg8X6MLSoAAEQ AIHoE5A6r0+jqze0KHO1HbrOHevYHqc1nN+doSShNYRk6Dt0mTvWVbEWlRbH7mQlCa1hh1BAAARA AARiS0CK1ityOyfPBMJRFB464z3gcpMUpYKuy9MSXkMyCg9N8kZKpWAToSbgEK8gAAIgAAKxIiBF 68NiSFYoQ9bbw2tIeFV4TZhnVIAACIAACESRwALX66MYAVyBAAiAAAjEmsCi5vWxDg7+QSChCHyY /TVJ8Xxl4IIkexiDQOwIYF4fO7bwDAIgAAKJQgBanyiZQBwgAAIgEDsC0PrYsYVnEAABEEgUAtD6 RMkE4gABEACB2BGA1seOLTyDAAiAQKIQgNYnSiYQBwiAAAjEjgC0PnZs4RkEQAAEEoUAtD5RMoE4 QAAEQCB2BKD1sWMLzyAAAiCQKASg9YmSCcQBAiAAArEjAK2PHVt4BgEQAIFEIQCtT5RMIA4QAAEQ iB0BaH3s2MIzCIAACCQKAWh9omQCcYAACIBA7AhA62PHFp5BAARAIFEIQOsTJROIAwRAAARiRwBa Hzu28AwCIAACiUIAWp8omUAcIAACIBA7AtD62LGFZxAAARBIFAL4vdlEycQSx3Hs7xql9lj5Jwek NoE9CIBAghCA1idIIhAGCIDAoghI/eV32tkN9ePv0PpFnV6J0PjHu56XGsYz7U9JbQJ7EAABWRPA er2s04fgQQAEQEAUAWi9KEwwAgEQAAFZE4DWyzp9CB4EQAAERBGA1ovCBCMQAAEQkDUBaL2s04fg QQAEQEAUAWi9KEwwAgEQAAFZE4DWyzp9CB4EQAAEeTFZ0gAAIABJREFURBGA1ovCBCMQAAEQkDUB aL2s04fgQQAEQEAUAWi9KEwwAgEQAAFZE4DWyzp9CB4EQAAERBGA1ovCBCMQAAEQkDUBfPeZrNOH 4EFgqQngu/aWmniU+sO8Pkog4QYEQAAEEpgAtD6Bk4PQQAAEQCBKBLCGIxYkfshJLCnYgQAIJB4B zOsTLyeICARAAASiTQBaH22i8AcCIAACiUcAWp94OUFEIAACIBBtAtD6aBOFPxAAARBIPALQ+sTL CSICARAAgWgTuBE/hyP1H4M80/5UtLHDHwiAAAgsKQHM65cUNzoDARAAgbgQuBHn9XEBjU5B4EYm gH+eEvfsY14f9xQgABAAARCIOQFofcwRowMQAAEQiDsBaH3cU4AAQAAEQCDmBKD1MUeMDkAABEAg 7gSg9XFPAQIAARAAgZgTgNbHHDE6AAEQAIG4E4DWxz0FCAAEQAAEYk4AWh9zxOgABEAABOJOAFof 9xQgABAAARCIOQFofcwRowMQAAEQiDsBaH3cU4AAQAAEQCDmBKD1MUeMDkAABEAg7gSg9XFPAQIA ARAAgZgTgNbHHDE6AAEQAIG4E4DWxz0FCAAEQAAEYk4AWh9zxOgABEAABOJOAL9VEvcUTAfwYfbX pnfEbX1l4II4Q1iBAAjc0AQWrPWes3XfP9ZlIyTtyhVn1aumPZuVw+2N3y+q6CKk2mCu3ZNPXbuG 25/+ZlGLlairDcdr96xNjlBzQ+PH4EEABORJQOovbVX+yYH4DnShWu8bMda0Pmoe2H6b9+pVsnpd qm/09MaiioaOAf3K93fnFaSutz+7zfn0xqJPGzpG1CubM/OKVq6/+NTa0JpD2+M7fvQOAiAAAjcC gYVqvf0jK6n+87vSkq+mZG9eQ730nzpOtG0HCjcTslmv4/Le+HnlyvdaSKXtQOFaOtO36Orz3rlc dFtIzcSh7atuBMwYIwhIJCB12kjdx33mKHGIiWL+413PSwrlmfanJNkniPECtX7i8qCV1GennyPE SojW7Pj7lGuEy0oXRrV+WzF5svdXGrq3QclXKdffS8gPh35Nq4JqPnIdyhcs5uRhNBojHuci1s5e OZuf2VuwIwtrtYCGUoezgC7YeBY6IqkcltlwKDepI5JKjE/Owv9I7U7qcJbszBEQ3DjDKSkpWXjW RbdcoNaT27a1dQ89tD0rmYw3a9K1r5S3MtX3Cv0qbllBrB4fURBC//JFcYuKXOG3wmsEi7n+zsbi w5ofzdUs7Jjg58cnng87MleF0GoB86zZwp6tM6nDoX5oF1KHI7RK2OHQ8KSOaGEJkpodGpjUBC2g C9qLUBI2QVKzQ4dDOSTscGh4Uke0ZOeb/1SI0ssCP3OZnLY2+x4q9LSsebBIZfV4mcx/fl2IauTN dlJ970pWdU3Qes9IbxcpzrwptOYuEZN6wSf+ggAIgAAILJjAArX+49aNmbcfGfUQ30R/8/6u0ju/ vOkBDan/Qc+Yj7gGX6ro0hZkKTc9oCI1+p4x+nmc116qINqc9WE1WKxfcObQEARAAATEE1jgGk7u gaGG/o2ZqVW0J67S8MaeLCXRdug6d6xLYTVaw/ndGYRk6Dt0mTvWMSNSaXHsTlaS0Bp2CAUEQAAE QCC2BBao9USRdeCVySde9LhJslIhOFEUHjrjPeBykxSlgq7Us5JReGiSr1IqZ60RLPEXBEAABEAg dgQWqvV8RMkKRch6e7JCGVJDwqvCa2I3PngGARAAARAgZFFaD4AgAALzErhBPr49LwcYxJfAAp/N xjdo9A4CIAACICCJALReEi4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQ IAACICCJALReEi4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJ ALReEi4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4Y gwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4YgwAIgIAs CUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4YgwAIgIAsCUDrZZk2 BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4YgwAIgIAsCUDrZZk2BA0CIAAC kghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXh gjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAI yJIAtF6WaUPQIAACICCJALReEi4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6W aUPQIAACICCJALReEi4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAAC ICCJALReEi4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALRe Ei4YgwAIgIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4YgwAI gIAsCUDrZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4YgwAIgIAsCUDr ZZk2BA0CIAACkghA6yXhgjEIgAAIyJIAtF6WaUPQIAACICCJALReEi4YgwAIgIAsCUDrZZk2BA0C IAACkggstda7htvLc5Jo0Rw+NeaTFCqMQQAEQAAEFkhgabXeN/z0xqJPH+8YGbFsqd9b9HzPAqNG MxAAARAAASkEllTrXQPnW0jl8QOFGRm51RadteqdCSmxwhYEQAAEQGCBBCaXsDitDYQ0OIUe7R2E qMz+nXmCWODY0AwEQAAE5EBgHgWMxuHkJefg8a/SK25RkSsie6cjFWkJMxAAARAAgXACS7qGQ7w0 gGuC1ntGertI8V3K8JBQAwIgAAIgEGUCS6r1yk0PqEiNvmeMENdrL1UQbc6qKA8H7kAABEAABCIQ SFri5ZHRziOZO6v4QCotjqO5EPsISUEVCIAACESZwFJrPQvf53G5iVKpiPJQ4A4EQAAEQGAWAvHQ +llCQTUIgAAIgECMCCzpen2MxgC3IAACIAACcxOQv9a7eguTcjrHhWGO1eUkFR7p9e+0H07KaXaF APD0FSZp+kJrg41cvTlJOb0u4hvvP322zxN8cJY9z8l9Oc39c/udpWlwdV9jYVKd/18Uj3XW0a+T 6PH/k7Oxw0lJjX0h//7M1VxIK+fpt+9ITlJdLx1P5+n2eb+awjfeW6dhX2RRuO9I3/hiv8jC1deY lFQXyE879XukV9jzdR5OKmzsCx49cfU3RshasJGr7wh1Q8c83t9+tpc+6p+z+MZOHdaw7+UoLD/d 7w9kzgZzHkyU8414Bk/mFIad3nPGHvFg3M83z/DZfYVCySk8eHqeUzniGGZUxv98I6T/NNUhWnLq TvUt9v0zY2iL3YzGh/Tj68NeS0h1h50G4R0xMRycju1MetsqCddgCQ3ObVERldUbWh2077WbO8wO 76TbqiNE5w46Fr7jtbYZqks52nODxRF+WGqNvZsOqFYYgqmSDai2m9+z0dGpLKH/+szZwJEG6zwx OgbM5iHHpNdCvZnnsbXTMZNKw5BtyEB798OUOogZ9uwfzRE+P5Mj/Hi42m7+8Ah1H07MaWkgXNPc MXodA93mIZpDlp/asBTP6JxuWhpUhKu12mxmNh71fKiCG0fYi/v5xsfkHdBSrKqm0NMhQsDzVMX7 fJsc0KuJ1jAwNGClZcA2T7jzHo73+ebopielyjRgs1kMNEWGobnP5XnHEzUDEjVP8XNkruWo2NP+ hwxaTlurJlwbO2GYlDRZnXaLUU2RU9XSNo1Q7G4L3eXUKr5O1dQ9Qk3tViOv1TRHld02qvED1erq 994zMP2mhaukAhHqZ3q8brNRrzfoSyMp17SV+C1HN+2XF0f6flbVVqu5yjba2kavXey9bTdW+wfU 0EGDdzbRoajUQqiVTd3szPLajbU0HFYqm8xUEwcM1ZUnz1EgfOEMA85Jh4WSYoXTdjAugeI0c4Tz Xw9sRnpY5L9tDrQPf2UXD/5i7DWU0vzQKKpZftilS21xey1Gem1jRdvQQeNwWpvotloYEEsHrfNa jTTHrKgq9Xx+DKWVxv9k2s0KpzWEYZkKw6kvVTUJAu+kYLnuRatjvM83OjSvUUtKdTq1uiEKk4v4 nm9sTsZVmgbsthG7c+4p2FRO596I7/nmNqhJpWlk0ut0OJ32kSFbdAY195BFHV0OWu8wUzHROSe9 BhWd4dratETbZp90tLFJ3C+YburNNq/bbqATIXpJELS+2mh3Oq0mqjJch32iiSPqJgvV+A6dipQa vWzuTzXOPWCqpm2sNofXHuYnFK+XOtFFY14/Oelgs1UqsSN6OsO1M02spBP7Di1R6we6qeiV6m1u r91KZw00eCedFRGu2mp32q0mOtjaDhubGpNSq8PrHumgFwLjiNvSwHE6s33ASMXOaGWtaRelerOb DpVxEW4j/ENyu4W3nLuNjl6ln3EdCB2zyH2zjqPjoVccNb0vcQxRhTax/NBLl/4X7CZGa7Y53XYL uwjQ8fBaX220Op02Ex0sqbWxyw/RWx2T7hHqqdQ4xA+wweG2m6o5rtpkc7jDsISE5u5o4C8MWuPi 5STu5xt/1Te4vVY1p4uC1sf5fGNnOy0qfvalbZrnLi0krxF343q+8XMvfkTsj9YQjQRFHKXkyi9N RSXfjVVbHuRI+wej1rYu1QPZa/P+UNvyr5bhC91E/e3fGr9gJeSj3tPHW37y/qeE1J+ni7tOwrVU F69RKrfurtQR6/lB1289wpn25/1xXcv1bc/ZXyzmvzhiNSGKzA13EHJH5tpV7ssR/AQTc18L3l/E 3qoHddzp3g/63zrDPX7/mrX3VpJj7wwPvtVCNNu//MFpK3F9dLrl+E+63ifE+lPLJ9cukdqWZ7au Ua7ZultfS2rOD6euzuRIK/etP2556/pzQ/biDPrxVjqclDWZd60mq+/atFbhvtxO/z3bR//R8nzL m8OUS41lxiq+QpHsG+s7nJNaVK+1/HPp4j8be/eDe8np3uFBs4n7zpZVWb+vJecsgxe6j6kfLRi/ 0EXIp72n9S0/efMzmp+f0lUmClJXXbxVqVy7+5kWGtvgr778CCFl3LfqWt7aph96sTiLpDC6yYo1 G+5YTVbeuXaVNwwLzfPMkpzxwDeNDVrSUndmWNzzl5mtg7fje775xtqLii5ZfrZHwf4ZuiIaX3IS 1/PN59vW1jbg8HZ20lvNppb9T9PnZIss8T/fSvVM4p1WbcveH3eGnIqLHNzCm0fjVFl471Fqqbz7 cVXXC8eOtqoffTGZKPL+kCs6cuTTLm3p0Apyjl5bc++++6br1zO/Z9RrbiLsWclqQSzoFhUyt/c3 dtf22h7/4O2e80078kyqJsfP7hMiY9/pMF3C/MQM3paHHrfmvfACd+mJ1kxClN+oVv3wyNEuUjmU seLcalJa6B+QyaAnq8h/EbIiJRDKCo44vMkZu3sdtg8uvH3+p0fy9nY1WRz+8fBj8bJRsf8LcznG hXzPoP/2KsbF72S8tzm9YH+lvtv52PaofIcFFUe1VfvCX5PSJ1op8LzvVBf98OinXdy3hzJZfkoL 77777us0QSZDEUm/Rj7i08LHSu9YCLk+mVHrdjz+wYWe8z/dkbeX3oG9cr//6FSCroVh8Vt4hhtr zz1cc2Bz7vas3G1NJ1IGPnOTrMVdv+J6vllbq6zEmpefQ6x0GkPSkt61uF/JXdyA4nm+JSvWZmZn rWLn3qrc+1XkuD9xi3iJ5/lGpypd9Cb2W6to/Mqt5Tou763LPypcu4jRRK+p5DuBhGxgbWJrz1rj EB+dnd7p02Kyeb1sAYQYB9hVdoCucdIVCX4NR93A1rXtFj09ahj6kN7e06UPasM/p1JZJtjzW7qI 4rTQ+8vaIWckP6Ec2DPSKK3h0BkBe6hAB8SeP9I1Hfa0hy7ktHn5B86k0sjG42ZP5/QD42zo6ga6 5O51WGhNqWFoxERfdXRdmw6ItqTL1fT5JKezTPJr8W0jdMWaPcyoNA4wN0N0Ub50+oml20r9VbfR nr1OVha/hEM74ReaaCKE51SObv5+vZo+bWDLEaSSPj6gNjQ/dJFKWMNpYM9RvPz6knZwlCaR6+DH w9ZquKaxwBemWmimq9tofng3M7FMhe2g61mqJvYAwjHURs+LBuuiF+zpM+H4nW9et8POisM+ZFJx tQOOaKwHx/N8M9NTu5Y9eaJPZejJoB1Y/CpbPM+3SbaCxOno+9FtY+e5zpwoqzjLYb2e6QRb5OVM I/7TxNJA9aqWiTeVeLYoLxS1acgprNez6SJfKvVsfdDWTQXBX2pNQ9SGaj37xIvDTFWTf4QY5od3 PuOPk67XN4R9SmaGgaRNXhyZuPPFaWbj4a9Gk86BWl4pWVi1Jno9Yuv1gcJVGthHdrwjDf5Hs9RI N+CetDaoVOwjSXa9lo2cPrJ2DpgCbkitiYm+UASpDfijMhmNp39+cWTizhcHfZ7M8Y/TaepM/mfE tK9aGqoQQCA/nN5CB+TtnjEemkT25dgc+3JsB3+1ppcIOp5gLP6e6Asd6RQhrfDsevrgArcS4Hyj V2mzWhX4hvAFjmOqWdzONxrBSMfUu09loE9lolH4i3F8zjc6e9IFTjh1bVsUZhbRAEJ93Bj/bpZ9 K4M3VakMrHQwKfN4XCRZOWO9kxklB9tMS56wFclPqM2S7HtcLm9K6szoI43H4w4eYVhoPpfLHeIm zGYpKniuKSFfm0HHGJQOZjTPV2tEwBII3+Px0AX+GekOHIjFa6TzBOcb/XaU2c83dijkHRqLzAg+ l+J8C3+Hxm484jzfGFovjgWsQAAEQGC5ElgOn8NZrrnBuEAABEAgWgSg9dEiCT8gAAIgkLgEoPWJ mxtEBgIgAALRIgCtjxZJ+AEBEACBxCUArU/c3CAyEAABEIgWAWh9tEjCDwiAAAgkLgFofeLmBpGB AAiAQLQIQOujRRJ+QAAEQCBxCUDrEzc3iAwEQAAEokUAWh8tkvADAiAAAolLAFqfuLlBZCAAAiAQ LQLQ+miRhB8QAAEQSFwC0PrEzQ0iAwEQAIFoEYDWR4sk/IAACIBA4hKY+Y3uiRtlTCP7H7d78vrU j9lF7uo3bkuLfAC1MSPgcc3zw7Bf+o0v3XTzTTHrH44jE/D9t8/nZb/kOUdZsWJFUlLSHAY4tPQE oPXk09p655mfzo1+k+XtL92cOrcNjkaRwOT/TP5dyTw/PZp+15r/c3xfFDuFKzEE3nvv4ptvvzW3 5R+Vfe/WW2+d2wZHl5gAtD4qwH3Dved73hu9aXVG7v2/u3nt9C89D/ecPtHSes56afWGAlXh7z2q Lc6YPhjetaezsfH6Iwd2LebHr31jZ4+/8FLn5dK/OV682f/b4J7RzsaX/5Os8PeouC397vtUu3Iz 6H7IIfrbyJ9fS3/8mfLFhBA+sPjU+MZ7fvbvH/zX9dUZ2Q8+lL9m6mT3jXe26vUvG6xXVnMFBTv/ 8H9/d9fWOdLiG+99/uh736pZFBPPaE/jsda3P8v6mxcPbfZ35us9dfz1X3oCaSG3pd91/yN/sHUN PezrPVnfe8ceeir40XmGTzae+IgUPHNo96r40Ixmr66xvn97/d0rRLkl7+vbt7LzUCgS3y+BZngV QQDr9SIgzWfS21i8saCo7NU2XUlR9rrUk30TfAtfT6Nm446SeteWg1VVmjxSU1GSmVre75rDnfc/ K6q6P3PPYTHvofHuF9QV9Rvy7vvN1JQpY++V96tqqgxv9w/T0t9/Yn9ZUV7m4bOj1CDoEDv80QcW +9WplvLd8I3W5aXvUO99te3lkqKC9JTDg8KakGf4cF76zrKqm1VPVFV9Z/3Vc2VFXP7h9jkWjNyf 9FbV7/9snnW+uUl5/qViR1Xn1bz77kyduuQQ93v/UFFT1c6o07S83b6/rIRL/27POHXl7j1ac8L6 hd+pb7guf2NZVf09RduXgdB7Bk+lrcsroW+Yl/fu4DILj/Tww1zA+2Vu5jgaRGD6vAuqxo54Ap7+ +gqTqqG788B2QkbrkjKPnv/osdx83/A/7aD1tW2vP7uLp7ynTLMtjSt79tSTZ8q3zuJe+ZTXS5IX lZRPPjYTTn/02ceCukihc0dV6+lXtgq+9X9+MCX73MXhH+3OICGHgprJeMdlNdVYiWHAuWezkopL avbesxef2Zy/qu///mm9lTSY7Qfy17Dh7SlX36cpqCh67XF38Sz3MsqtB7zeJxaXFuJyEtUTf/5s +eYQpqqmhubA+XC8/1Qat/fC5Ynta5JXrCarhek/vWg9vLHGyhkH3py6UQtxIqtdz2tH9xJSO+J9 NiOZ9DYWFlS0Dh7YvtG2gPeLrMYd72Axr198Bth8r7J4G+/oN9NVfofv/PM/EFLa8Iwg9KxSufVb xkq18trn0w+2fKPNhxv9801m4vqnPy5uZTN/16mDmsONjeU59BFXUlLOvlO9Y+w4PTDcebCQr0wq bO4cpjVUyDTlJ/t6TtLap58t4/Z3EWtZUlJ5X+gNxIzHy8mZOSpi5R3yf2Ycmq6U95Zt8F3CNRTx q1iKdVmcMBrf8On9JlJpekIQer4yf1+1llNdvzo9b/eMttc1dk7N9D3DZx/Oe4amhW5ocjR1Rw76 E1DeOOiH7Os/e0TIVZLmcO8Yy/DgqYPlze1nG/clJW1+7J5Umpau/dk55adC0zIDs/LOO2bsEbLi ZkLGjjycWdO1bISejs975WOiatJQoafljswcQj6l7EW9X4LoYEcagUVNIaV1tVytFblnJifp4Eb7 OtvP6elbuum5uwiZ+KC9i2hN2YGVWY+Hvv9TNUfPFM/k4P7keP2Jr1YfIH4z8pnZ5CmnlsmuS6b6 Y6ZKfcdQwS3n/rpgb0HrTSPu4pvfVW/ceaVSbzme98vXj6h3biRW5x7iMrXsN7WQSp3+D1WZt/Wf rDFpDW3fWx/6LNnUdqZ9/NabyE3Xf/GWoayL6J/7WiAW4RC5LuzfkvnQ9s1yPzM273llcg+9No72 vPVW25G9VlL9LzmriO+SmRDdd7f7R+fzeXw+osxrvtgZQMFevVesNRXkyQOFQlq8V0e6rBfpupb3 6n+ZrCaTVWXotube9OGfFZRkmz32i4eunn2GUx+rNXS0biVnnt1ZsM495D3qdV1q2V/UQq/3hvqv rvnvFQ+XtKhrdd/LD0lLV19XZ+f49evkJvLJ+SY64a1+hKPrNPwVwW5uLC+r6qLXprZlMaMXGCvL OyfL6QjH+rteb3u27JiqtjtXOdEs5v0yM0nYlkhA7u9oicONobnr3Vd+uP9YF+3BbncSkk43VLkb BL6uvufT8qr8nauaHJ3lM1ZdV0cOivqobjv6WCE9euCVkU9aM4+bPvjKir/tIqXmZ9QbUsmmfX9e e6L1+Jn39miYA123/dB2tiixWqOuuZT7zV35gcsHOyqUqpKiwCZ7vWx3EOIPJOiQWu/cHniqO7OB DLc99nf/tkpnYrcwl0c/9azz/ZxmSJPiT0tjXlpF4O5G3WA5cyA3MMQQeOzp6dTTjyaLaU8ufei9 9R+t+lSu7N+Hv+c6doxoDfu/cV+yj+yv09eYys5Za/gbPK3V3cw/9vX8XEU+LtLsyg88a+V7WkFv qFoqdrYEumWvV8ZsnqwMQg911dArMkdvwI6pf1zuPhp4ojvTWMbbtrd/8uxRupxGiOPyBPkqfRX3 fpHxkOMbOtZwFs3f53GxT4Iri492Tk66O3SqmqK/7Ofv/7v6LgnLNcot3x8ZGbHZbfQo7S+ZuJqF dZi0AivpKkjjlwQKG4VHuv6ArpDaovxAcGnCZDBF+ZuEtBasu52WtNuz6Xq01fLB58xI/dB9/Ooz nXtec9L96fWIgAv6VrK4JwPF2aFT1xT9qRBn8KHJyTOP+T++M91WflssKz6iyCo+c/HipNNaSVp3 Nr+tSL+LJuCLz4XlGWVZG03LiM1mreUIo+bq0/CpSMurIKTidn9a+oIHr75/ix+PIruglJBf2of7 6AWkZW96WhpNSzpXRu35j9bQFfptm/xXDe81WnstNC00V6oGayApk16HRUtadhx7g9nS/4nW7LjY aTNx5Fh2xSkhaFYt7+Jz8anZXPwjmhm7uYkc2/tiHzuLZ3+/yHvACRI9tH6xiXBZG9PS8vv8b0TF fQ/Rabbtqlf54B+Vkhb1mWH+gGJVRkbG2jUp79MbVVaUe1612aj2D5gI4UwDdMtme3XfjMk+nZ+T mjfeDwTHLhlXPF6v61P6UMsW0IYRs6mt5hsrmZEzoroHmguvdA45VZRfuzePtgqsUc88NGUj743X n05LLTvtH4My+z6qyh/8wqVcr+FIzUG98PRDuZamJWOt99JpK2EIlFteZGlxDLRV07uqAbuQli3B IEz9I/5k+2zvt9I7uFt/O1dFOF23Py1ee4fR9OAGJVV3bhcXcoMQ7Cp0L3nVllyOkEufsBUcehlo ejKfnhNrd581VdNryfdP9oc2kOO+y/JAWlrtO/6JzZq8+/kboFvnfL/IcZwJFzO0frEpUa6/lyPW p2tPDY6NDfedfYJOCTnNXUqy+ds/pLPFko35ze19Y+Pjo4O9R/alV1Cp53VVuWbt2rVr1mbeqSKr 78ykW3QvSOrZ3X3NjrqzfeMTY2ePfL+GkCeK7s7+xvdo7RN1Z0cnJkZ7T2UWqJsG/R/QDJ0xRhiW 6d03evt6e3t7etpPN5ftrCHqRwMz1MAhepTjb444AAADhklEQVQd7x1lYiPvkvMIncqXHKEAx8d6 TtXsbSXa0u1Kskb7qoFYq9Zp6noGR8fGx/o7TxZmqtlKApuKK4REZK67gz41XLdG2AuV673ZFZ00 2aO9NbtL6FP57dm/df931NaqHY3t/TRZPSee2lmiHvGnJYjhlaA9/07XJUtvH0sLLacbK/ZbSeX3 VP4bh8B9QMbuGhMdTRnX3B907xfJX8LXKbccVJP6HX9GGVL8zX9cSt8Tm1enzv1+SfhRySBArNcv OklrCtu6m0p37M2u512pqy0vPsHWU5Iznu0dSa+u2F+Ut1/oRK1rM10rOjazxxQq6fRjllPPZqng CJNGenevrq7uUudRlael2mAt30ovBrtHOhp271Rn8rXqauOLe7JI/3l6AZlaTU6hV4nVM7uY3t5f VDC1o9I2DDzH1moEVZ95iNo0WJwH2JK0jEvG7to23WdFav9zklJd23PFbK1csXmPY+DWP/uOeke2 gJZoG4x/9EndP80YK7twcvwyDF+Zwj6Wyl+i+d1S7ac7s9cJm21DtVn0PVSuN9nL1EUcXfqhLXVt Q7vWJodMwqmLDSumssS3pjcS1M2xsoLpU4KrbOqupx+EJS52sZ8uit0/HqjuzN7P1T7oPSrz5+bK x/SWj76VtzNbeEyhauoe2c3+haGY98s0EWxJJZBEbzyltllm9s6f/Zun///NPag1B/8k6abQN2pw E7oK6U5JSVUoQi+fdD1/wuVJTlauWhU6Qwz2MHPP1ZiTRlrdB7YmM7epymCvrC8SVjuzvdy36Wn5 xovdc49C+eVb7/1m3tw2xONxeb2pqcrwT8e7Jibo0lhy6iql6LS4+pvTuDar98xWmgEvSVUqZybb 5/G4vd4UpXh/88SemIcv//KXl0YuzR1b/u/8zoqbpi+WEY3pAxUviUBrQe+XiD2gMogAtD4IR8Ls uBqT0r4wO55l67UoiULA1deYlnfC7LyYL+97nkThiTiWksDMeclS9ou+5iaQkm/UXw/7hPzcbXA0 1gRSN6j0TZlIS6w5w38sCGBeHwuq8AkCIAACiUUAn8NJrHwgGhAAARCIBQFofSyowicIgAAIJBYB aH1i5QPRgAAIgEAsCEDrY0EVPkEABEAgsQhA6xMrH4gGBEAABGJBAFofC6rwCQIgAAKJRQBan1j5 QDQgAAIgEAsC0PpYUIVPEAABEEgsAv8f7GAqAdNxMXUAAAAASUVORK5CYII= --Boundary_(ID_UQnfOsK19lMNJYbtgzxRtQ) Content-id: <7CDBAAAB-946E-45DC-A4FF-E7D42EEEA394@gateway.2wire.net> Content-type: image/png; x-unix-mode=0644; name=bk_be.tiff.png Content-transfer-encoding: base64 Content-disposition: inline; filename=bk_be.tiff.png iVBORw0KGgoAAAANSUhEUgAAAgYAAAG6CAYAAABz344zAAAABmJLR0QA/wD/AP+gvaeTAAAACXBI WXMAAAsSAAALEgHS3X78AAAACXZwQWcAAAIGAAABugBL/GM4AACAAElEQVR42uzdeXzU9Z348ddc yeSYnCTcJBCOgBAUUEAFDVor1Rqs0kOj1bYbrNtV7KEbt3W3tL+6cbdHXNeibotbwdZiu8YLPAAF lSAGIRwJkEAgJEBCrskxk8zx+f0xySRz5SLjhOT9fDzygPnMzHc+853vfL/v+Xzen89Ho5RSCCGE EEIA2lBXQAghhBDDhwQGQgghhHCTwEAIIYQQbhIYCCGEEMJNAgMhhBBCuElgIIQQQgg3CQyEEEII 4SaBgRBCCCHcJDAQQgghhJsEBkIIIYRwk8BACCGEEG76UFdgNDu1/UX+9HEN4eG+97W3G7np27dS 8pdNVJLGvT+8i5RQf1r2U7z4mz8FsT52rM0WMERgNPZv4/baUt5+611Ka2DStLlc8+UVpJi8H2Vl 14tP80ElXHPXQ6xIM4Z4R4K9dg/Prn+XpvA07n3oLlK8qnR6+4usf7+GG7/zECumD119rae28/Rz 75N843e5b0Wa3zpZPQ5II8mTUlnypZtITzIGqLeVXS8+zwc1ViyNsdz7xBrSjX1tF2hvJ/n6e1lh 28H6909z43ceHdL3OmxYq9n+5icY5t7AsvR4n7v7OhZCoXrXy/zhg3Ji027iwbsWe14o7NW8/Js/ UM7Q1/fU9hd57v1KbvzuT7y+p53f4RrwODpjk5l75U2sWDChR73P+J5T29tJvuYun2M+0OPb25O5 65Gb2PPsHyhnMvc+dN+w+Fy+MEqETFFepgIC/q3bvkNlgYJMVWgOdW2VUuaioNbHXLze9d4z81V9 Px5vqyhQGT77LUvtrLF5b1mtz3Tdn1/Uny0Hn7kov7O+/vdlUd48Baif7x7a+na9bmZ+US918veX pXbW+K930fps9+NyNhQPbLvrClXR725TgPrdMPlshpZN7czPUoDKyCv0+4i+joVQKOqsMxl+vovm Qvf3bucQ17frdX2/p93fYX9/uVuqPOvt58/f/g/8+AxVWB3c891wFurfoKPauBvXsmHD94gOa6bg 7gfYCGSvW09WmomWjjCWpEdxDICJREYA2LHb9ehD9akZwBTE+kQYOsP2mBgi+ny0nbd/9zOKATJz 2fLTuWy64W42UsC//rmY7Q8t8Hh0eEyI9lkgXe+VGAx+7p7zD1s59U07puT4gWw18N6y29Hr9URE xnS+am91ymD9tg1cnQQnP3mJrAd+BxTw+YkGLo/srndkBFS//jgLH9gIQPb6Ip67b14v24Xc9ZuY 29mi09HRwbiF0zB8PNw+nKFg5+Drz/LU7x5m4w5XyVSjwf9D+zgWBvq6VjsY/Xwp7XYr6I39aiY2 hHd+SInhvo83LeSDqgrMRDLR1I+N9afWncdnZEzgDXZ9h9du2sZ35iXRVlvEMzfcz0ag8Ng5uHlC d73JJH/T9xjT9eSODqJnTw78Pskkf9ODjKGj8+HRjIvoOt8NxedyiQl1ZCKUUsqmNmS5ItX1xZbu YkuRyu6MYNeuW+uO0nPytymLUspSslllZWaq3PUFatO6LAVZqtCslKWqUOXldLdGZOXkq6Iam1LK orasy1GZmdlqc4krBLaUFajszEyVnVugXK9sVtvWr+3+BbguV+VkZ6ns3M3K3Ed9lKVErcvKVNlr 16n8ddnuyDt3w87u+mZkqtxNJZ1v0Kw252arzKxctXf/ZpXZM7rPzFXF9RVqw7pclZu7Tm2rsHjt s3qVl+F67IYS130Vm131zvL5NWx279/ALQY1akt+93siM0dt2lnh2kclm1V2ZpZauy5frcvJcP+C 3lBY1f3swk0qq7M+GTl5akN+rsrKzFaby/z/1HC3jpClivw85HjBz9Vt12erzYeOqLzsLJWZlaO2 Vdk6n7tJZWVmqux1W5RFKWWu2KlyszPd+3tt/pbuX3n1xSqv676MbJW3LifAPupZp2xV4i6tV+s6 31deYb3HYzYXrO8+xvJ2Kpvyz/92uxX/PtujxSDg8WurUPnZWSozM0dtqXAdzwW52SozM0tt6Hxu yaZclZnpOsZc2+n+RZiRtVZtKfH9/It7PMf1dSzr3Oe5qthcr7bk53YfmxmZat3mooDv1eP49PoV 6m+f9+dYUEqpsm0bVHZG96/Z3PVdn7HrO52Vla1y89a5H5ORtU4V19t8jwGyVV7+OpWd1fN76Ocz 6WoFylyvfKpkKVN5OdkqKztP7d/vOgdlrd3gPuYK16/t/kxsVWpzXo57H2Rmr+s8F7lUFW7w+N7k 5wZuMXCfI0tsPerpenxmZ2uAu95Zm/r8hPp8vPt81/25BPocus7F7v1gLla5WZ3HkEUppWrU+pye x+7wJoHBsBDgwtXjQgwZKie7+2S5vtjc44TS3dxbdLqws/kLlZG9Vq3NynCfEIotvk3q3c3361W9 UqpkQ/cFPWdtTo8L5XpV30d9ursaOi9Qud0nhOwNxe7X6j5BmtX6LFe9P9y/xeMLl52Tp0pqu5ss 8wq9ThSWCrUpP1/l529WVZ3b2rw2I0AzeV+BgUVtzuneh7lru5vG83ZWee7nzGyV7W7SzFElNqVs VVt6nPjW9rg/cCDS18Wg+Pddzes1qqDzfWV3NtNvW+e6nbO5TNmqtrj3UfbaXJWT6bovI3ebsqka d/BERrZa2+Mi2XtggMrJzVN5eXkq132BXqeqvB7T/ZerqlRg3c/JULnrN6lNmzapTZs2qA2bC5VF eQUG9b0dvzZVsLbzvW4qU0qVuI/HjNydSimzyu/c9+uLTrs/88zc9WrzhnXuOnjv75qdeZ37KE/V K6XMRZ31zVqvijZ1HsOZa9Wmgg0qp3N/rt3S2zvu3G5ZiSopK1GbczMvKjCo2rau+3u0dm33d2xt gbJ4NbFn5WS7jwdX03mNe5+QkaXW9gy4AtRHqT4CA/f3PEsV1hS7P4OCCqWUKlM5nft5W01N5/cb lZGVo3LXZnV/byxKWco2d39vctaqnF6/N93f4aycdSovL0+ty+0+P23o/KHjrjcZKid3ncrNzVW5 ublq7dpcPz8uej4+U+Vv2uw6NjdsUAWFVT6BQa+fg/uxGWpLvVKWHt1n+cUWpWoKOm+vVWXDPy6Q wGB46CswyFTbapRSyua+SOQX9fz1lqk2FdcopboP9Iycze4WgA3ZnSf7zQd8XqdnYGDucZLJ21nT ef8G90nS3Ed9VI8vx6Yy16vXbMtzb7/aT2CwIav7i2cr6XqtDa6622rUlk0b1IYNmzx+ZfioL1Z5 2d0XEN9rcR+Bgfvimq264o+yztaHnvUmM6/z10Bh5y9IVwtNcWcwlbG2oPOXZJnKzeh6vRpVsq3z hLNpk9q0YZMqrLL0IzDovli6+6Az1yuzqlBrOz+DnfU9fjHlblYVNfWqomhD53vJUp98tsl9Iu5q iCrMz+5XYOD7l6k2l5gDPsZfbkHf23Xtz57vtffjt0zVd13Eszerqh4XFjLyVEXNzs7PZa0qs3Uf y5lr16vC4iK1ZfNmta2wWNV7H0q2EvfFbEu9RRV25v6sLahw718yc9TmnUWqcMtmVbBtpyqp6n+n c1ewPbjAoPt9rOsKRmq2uY+/InP3sZ3T2QJQX9h9vNS7H5vj3vbOvKzu+tiq1JauY3PTJrVp0xZV ZeurxcDzgrktN8MdrFnKOr/D2Ru6vzdkq21lNaq+ptij9alr32bkbunccIX7/t4CA7/H3ybXvi3u ke/i/Zfn57sf8PHrCr3eZ1+fQ/e5MHdbhUfeTfaGYlWxJdd1LK7b2e/jJpRkuOIlIYY4I4CeqfMz fO/OeZCvz0vyLJs0FlcSrYklN2QDcLzaTLv7Af57Gl39eFnceKVre6bZC8keaH1YwoLOFN6k5Std z9+xj9NdLx7e1WPXTOXJ7mdZbJ0PMLdjAdAncfNd93HffXexIMl/fcu3Po0mIYPHNhaTkZ1Pifkl FvTaLe+7nYbKYleuQsaVzOl8bsqV13S9VbfMm68jHsA0jZszXWUGgPZm142EsZ1bTyCuuOtZLbx+ w2ruvvtu19/9d/NupYWBMGXcwlqAHVt4+eVX2A6Q/T2ujLdydJ9rB+54cjWpyQmkLryfrpe2dW0g aykzOjOq52de249XzKKwXqGUQikzW3IzgR2s/u0OHD0elZ1fQEFeFgDP35/By6XNfWw3k81FZVSU lVFWVkZZxf0E7FH2e/w2EX9l5/G0sYCNLxcAGeRkZ0LxVv60fiM7gIx1XyNNb+Lr+ZvJycxgx+8e YEnGQlauXs0v//gJZrv3IZFO9rpMoJj/++vfKdi6A8jim8tSmHfXv5O/Ngt2PM/q5QtZsnI1WY88 z7EBfIS29v4/1p+u7+Qt17sy70m6mnuzAArYW9HiftyCeRNdb6crDyQGrJWH2NF5DMzp3NlXrlzZ vXFLJSu7js277+buu1dS3vO99SP948pvPQjAxi1/4y8vvwZA7t03Ee1+xEZumJ5MQnIGT3QenFs/ PEZXD/537+z8rpHCqgez+ny9/ML6zmNTUbEtD4Dn717Hvp6HX0YeFRYLZrPZ/ffDXk8MaymsqHAd l2UlVDwwf4Cfg4VlX7sbgLde+ROvbtkImdlkZ8DG1/7Mn/7vLQC+ccvlgz0MvlASGFxq/JxkshbM 6nG56/x2WFo7b9s5dMD1bZwYY6Ar18ba6jo72tua/Gy+gOf+sg+A8rdfYyOAuf/1gUIqajpfp6TI 9fzMy5jQeb4q+OggdgDrOYqK/Tw9Jrwz+dDKqdJSDh4spdbq+7BTWx9n+sqHAcjbUsKBlx4iva9k KJsVV9Jk959pXLrrvuLDVHS+TkN5ies/5na6riMx7uQxI8k9TphdCUzFRXuotoO9+nMOuu+N4x8q yigpKXH/fXeOZyVbrXboUR/v6xb6NL6ZlwkU8MDdj1EM5N5/PUaMTL5squsYyC/EpmxYLDXsLNjM ps0PMSm8o3OHH6Cqc6MVB/fRHwb3AdUj7ep4Bd3n3iweybmN2x7NZ11nbHj37Ccotfa21RimzUkj JS2NtLQ00lKS/IRpvR+/GGezem0GsJHHntgIWQ/y6KPfA3bwxBPPA/DgLZeDvZZjVdHc/ugzVNXX U1a4mSxgx/MP8MqBBp9XvXLVvQA8/8DdPLkDyMlmfjycOl5LauYDFFfVU1NWxPq1mVC8kayn3sHa XEvpwYOUllf7fmaD5O9YaDcDFPBRV72tR9hWAJDJ5andl992m1ctzBA99QoyAQr+xJvlzUAtf3nm 2e7HmOZQ0ePYLCmp4Mqeh6e5HSu462O1+75T07ybyQXY+Bj3P1EAZJF1zYTuiChjHWUWGzabhbLC LWzatImf3zEHW2dA/dpbezu31MCOVwr63kndBycGQ/cYQlvPxyQaiTQaiYiIcP/1+iFlpTMnJcV1 XKalk5LkOzax98/B5A5ai59/gicLIOd7j/KDB7Oh4EmeeL4YWMuNGUOUrRlsoW6yEEr13ZWQ5R4u 09X01bMroWcTpUef87p8dzKPd9MzZKm8/FyPHAKzUqqiINd/05pHV4L/+njmRGSr/PV57n64zLyd ytKzD3tdnlrr7lN0NUn27IvO6swx6Er68s0xKO7xWqiMrCyVkZGhMjJQGTmbvJo/e2mGzFyv6lV9 dz9tVq5av36de7+sLSjzmxvRswtE1e/0SJzs+dd3joHvX2Z+kW9CXsmmHo/JdjcLu7t6yFS5+Ru6 P++MPFWjKlQuPd5X/tp+5xiQkeHanx7NtWUezd5dx4CtZ5O+uwvA33az/A776vle+zp+leqRE0BX 0llXv3aPPlxbd1l23ma1c9sm97GYX+yvG6DHvgK1ztVXpgrzOruostapLTu3qfzOPvrM/KLuLp6M 3ofXdn1H+u5K8H8slHTlOZCp8tav7/7eZG1wdf/10T3Y1RXj/devHAM/f+s/2dmdY9C5Kwt7DvvL 6exSs5W592nW2ny1ocf3qqDKpixdXYedn3XPhNNeuxK6js2MHvXKzFc1fdSbLN9ukZ5dJj6foaXI 4332/jko1bNrFVAbSmweuQYZ67apS4UEBsNC95d3vd/AIFsVdZ5tuw7k9T0Dg/WeX/Caok09kgBd iWebi13btVXt9Lig5qzrzLjucXAXF6xXa3NyVHbOOrW5YEPn/T2TD/3XxyMw6PGlzchZr1wJ9fVq U073FycjK6czI7nrQlejNrhPDpmqsKbYfXJf730y9whCvE8AG/ofGHSdLMwlPfIUOoOCDYXKppSf /dy1vR4X6LJtKm/tWpWTnaPyNm12bytQYNDzpOhzwl7fHRj83v38KpXXeSLKyN3ikRVftiXfcz6H rFxV2JmTYSnb4hlAZeD3mOmtThmZWSpvU2HnyJKux2R79IcXb+hONM3dUhFgu9m95lN0vdfejl/X BruDwvzCeqWUTW3uPK4ye5x8zSUFHgltgFq7fqfyTUFz6e4X7s7JUJYKtX6t51j3zOx8VdIzkPVz wfF4f505Btnr/edh9HUsKGXxGCnUVQdXGo9ZbXKfO1y16M4L6vwe2KrU5rxclZOTrdau26AKNuX2 HRhs6CUw2Lu7+zzQIzh0Jwp35ie5Xnpnjx8Aru91/rbu46N4c67HfVnu5NH+5RhkZGSq7NwNqmvw T2/1JnuDb2DQ9Xg/93UfZ13vs7fPwaU7aO08hnokZK+/hObpkMBgxLIpc329qq83+72vvr5e1Zt9 E/p2rsv0uChWbOlOHuwz3apH4FBsUcpmrlf1Zt/TsMUcqF6h56pbvbIMIHPYPWlQ50nCVlPYGdB0 /8r9Amre+Zn6u+wF/ryHr96O3wHuGbNZ1debB/SZ+tTG4tq/5ovZyEXvEsvAP8eui1tGVmdw1R2c Z/cyXHGodX2v/NbcZh7wdy6kBvM5XGI0SinVV3eDGD2aS1/l2tmr8e7631BUz30L+phsp3kPK2KW sIMMCs0HWHyJdKddNPspnrozlce8ukez8nfy6kPLZN5xEVL7XlzDwvuf9yzMWEfZnp8xDGYHF8OQ BAbCV3M1e/YWU1nfQlj0ZOZfs5AUU38ub82UHjxOG7HMmZfG6DrnWCnft5fDJ87TERbNtMsWsiAt 6eI3K8QQaDh1kM8PnKS+AxImz+TKxemMlrhdDJwEBkIIIYRwk+GKQgghhHCTwEAIIYQQbhIYCCGE EMJNAgMhhBBCuElgIIQQQgg3CQyEEEII4SaBgRBCCCHcghAY2Nn38uNoNBo0mhU89Wr3OnPlW59m hUaDRqPh8Zf3uBe7ClTeXL6VNfNd5asef5lqe+/lNJfy9JoVrtde9Th7aodqzTMhhBBidBjy2Vob 9j3LwrsLKSiuYL7tY1IXZhBbbOa7Me8wfeXD5G8rYUPcIW5buISIyTXkTvnQb/nPlpr58fSVnM/f RkVWHM+lLmRl3GQO/HCC//JHr+TVB2fzcHM+JVUvcOiZr7EkOY4a9Sgy/5wQQgjRP0M+8+HBF9fw FD/gpfvmAQ08pUmAIjMr92aRse9R1HM3A7DvqfksbH+e4uRcv+XmVfuJySilSv2WCUDzvqeIWQin i2OZ4qe83nIjd0QsZG2V4rYJgHUP8yNyeKb+AMviB/tuhBBCiNFlyFsM5t33HC8Bp7Y+zf0rH2YH OZRlmGj6CDLSkt2Pm7z0TvjBHhq/67+8ZRXAVPd83qbJi4BfUta6ym/5seYbgUzGdt1hnMadFLP7 ZAPL4gceGWzevDnIu14IIYQIntWrVw/qeUFb+C1mxrU8ujmfutUP88zbj/IdoBib+35jVDgUW9EG KLdjBKzufAOMUWRS13kjUHldjxroic2AJhuDMtgdKoQQQlzKhjj50Mr2p3/B66XNxKct4OY7H+KV /Ex+V3rBdelv7HA/suKjrZC7iIgA5XE2gHZ3AGCt2MMO7iQ1zH/5tM6l/Frdd5zktWK4Zpr0Iwgh hBD9NcSBgZE466tkPegaKWBvLmXzH3aQaTQw49pV8OS/sqvaDs2l/PHhHeQsSQtYbppxLZk8wYZd 1UAzb/7xYciZz+QA5fGmqXwjo5hHXtiFHSh98yV2sBZZ+VYIIYQYADXUzCVqXSYKXH8ZOetVhUUp pSxqW15Wj/JNqkb1Vq5UxbY8dzmsVUX1vZdbKraoLHd5htpQVDPQ2gshhBCj2pCPSuhit1qxo8do 1HuVN2PBgMlo7Fc5divNFjCZ+lmOneZmCwaTCe97hBBCCNG7oAUGQgghhLj0yJTIQgghhHCTwEAI IYQQbhIYCCGEEMJNAgMhhBBCuElgIIQQQgg3CQyEEEII4SaBgRBCCCHcJDAQQgghhJsEBkIIIYRw k8BACCGEEG4SGAghhBDCTQIDIYQQQrhJYCCEEEIINwkMhBBCCOEmgYEQQggh3CQwEEIIIYSbBAZC CCGEcJPAQAghhBBuEhgIIYQQwk0CAyGEEEK4SWAghBBCCDcJDIQQQgjhJoGBEEIIIdwkMBBCCCGE mwQGQgghhHCTwEAIIYQQbhIYCCGEEMJNAgMhhBBCuElgIIQQQgg3CQyEEEII4SaBgRBCCCHc9KGu wHBVU9/Cz37/LkqBQmEM0/Nfj2aFulpCCCFEUElgEECHzcHJ6gb37TCDLtRVEkIIIYJOuhICcCrl cdvucIa6SkIIIUTQSWAQgNPpGRgor0BBCCGEGIkkMAjAu8VAKXBIq4EQQogRTgKDALxbDEC6E4QQ Qox8EhgE4K/rQAIDIYQQI50EBgE4/QQGDqfkGQghhBjZJDAIwF9XguQYCCGEGOkkMAhAuhKEEEKM RhIYBCDJh0IIIUYjCQwC8Jdj4HRKYCCEEGJkk8AgAKd0JQghhBiFJDAIQElXghBCiFFIAoMA/I1M lFEJQgghRjoJDALwO1xR5jEQQggxwklgEIAMVxRCCDEaSWAQgCQfCiGEGI0kMAhAZj4UQggxGklg EID/tRIkMBBCCDGySWAQgMx8KIQQYjSSwCAAf8mHDoeMShBCCDGySWAQgCQfCiGEGI0kMAjA/zwG EhgIIYQY2SQwCEDmMRBCCDEaSWAQgEyJLIQQYjSSwCAAf0ssS4uBEEKIkU4CgwCU3xYDGZUghBBi ZJPAIAB/CybZJflQCCHECCeBQQD+5zGQwEAIIcTIJoFBADKPgRBCiNFIAoMAZBElIYQQo5EEBgFI V4IQQojRSAKDAPx2JUjyoRBCiBFOAoMAZHVFIYQQo5EEBgHI6opCCCFGIwkMApDkQyGEEKORBAYB yHBFIYQQo5EEBgH4mxJZkg+FEEKMdBIYBODwEwQ4JcdACCHECCeBQQB+WwykK0EIIcQIJ4FBAJJj IIQQYjSSwCAAGZUghBBiNJLAIAC/8xhI8qEQQogRTgKDAGTmQyGEEKNREAIDO/te/gXzNRo0mhX8 4tV92Hstt/L6L+5hxYoVrFixivnzV/ByaTMAzeVbWTNfg0ajYdXjL1PtekLAcppLeXrNCjQaDZpV j7On1j7QyrtJjoEQQojRSD/UG6zd9SQL736VTYVlLGYP05csJHxnDfez3m/5o0vr2PzERu4tLGFZ rI22NkicGAH2cn48fSXn87dRkRXHc6kLWRk3mQM/nOC//NErefXB2TzcnE9J1QsceuZrLEmOo0Y9 StIg3oesriiEEGI00ih/V8CLsO+p+SxkA+rRBQBsXaPhsbRCNpDjt/xA9nnmTyzklZofENFmYGJK Enqg+eBzxGSUUqV+ywSged9TxCyE08WxTPFTXm+5kTsiFrK2SnHbBMC6h/kROTxTf4Bl8QN/H8+8 8gmv7yzxKJs2MYH1j98ego9JCCGE+GIMeYvBgh/uwaY3AmA/tZWVz8P6kjksmO6/vKHyQ4p5ktnJ bwHFQA6F9f/NHNqBqZg6t2uavAj4JWWtq/yWH2u+EchkbNcdxmncSTG7TzawLH7gkYF0JQghhBiN hjwwQG9Ej5V9L69j4d1PkrOhiDXpps4X8y1vKF3Klp1l3LgsDT21PLcqmZyX1vBRJoAVd5aAMYpM 6jpvBCqv61kRYjOgyTa4t1FeXu5T1thkZvPmzUO+y4QQQoihtnr16kE9b+gDg86L+wMFa9lZYWZZ iqnXcn3MBGaPTeusSBLXrczkgSYb2ADa3QGAtWIPO7iTP4T5L5/maoyg1X3HSV4rhp9OG0Q/ApCS OpWS6mMeZRERkYPe0UIIIcSlYMhHJZS++A88ULCOMvN/sHQsNDc0Y7UHLj++cTqpCU9xygr2hoM8 98AOsieNwTTjWjJ5gg27qoFm3vzjw5Azn8kByuNNU/lGRjGPvLALO1D65kvsYC1pg8k8xP+UyDKP gRBCiJFuiFsMmil8rQAoYHrME+7S/KLTxPgtr+ehh8rIPzid1IjHAMhYu4kP7koDYMO2PFKXT8R1 z1qK6m9DbwpQDnz79S1sSV2O4TGADDYUvU/KIN+J008QIDkGQgghRrohH5UwWHarFQt6TEa99x00 W8BkMvavHDvNzRYMJhNGBi/vfz9k26dlHmWmyDD+9h/3hHpXCSGEEEEThByDQVbEaHSPNPC6A5Np AOXoMfm/Y0CU31EJwyKGEkIIIYJGpkQOwO8iSpJjIIQQYoSTwCAAmcdACCHEaCSBQQD+uhL8tSII IYQQI4kEBgEECgKk1UAIIcRIJoFBAM4AgzVkISUhhBAjmQQGAUiLgRBCiNFIAoMApMVACCHEaCSB QQCB5n2SFgMhhBAjmQQGAQTqSnDIyAQhhBAjmAQGAQS6/kuLgRBCiJFMAoMAlFNyDIQQQow+EhgE 4JQcAyGEEKOQBAYBBByVIOslCCGEGMEkMAjAKV0JQgghRiEJDAKQ4YpCCCFGIwkMApCZD4UQQoxG EhgEEDjHQOYxEEIIMXJJYBCAkimRhRBCjEISGAQgXQlCCCFGIwkMApB5DIQQQoxGEhgEEKjFwCk5 BkIIIUYwCQwCkBYDIYQQo5EEBgHIPAZCCCFGIwkMAgg087GMShBCCDGSSWAQgKyVIIQQYjSSwCAA JcMVhRBCjEISGAQgyYdCCCFGIwkMAnDKzIdCCCFGIQkMAgjUlSBrJQghhBjJJDAIwCFdCUIIIUYh CQwCkORDIYQQo5EEBgFIjoEQQojRSAKDAGQeAyGEEKORBAYBSFeCEEKI0UgCgwACDT6QrgQhhBAj mQQGATgDdBnYHTJcUQghxMglgUEASloMhBBCjEISGASg0fgvt0vyoRBCiBFMAgM/lFIBZziUFgMh hBAjmQQGfgQaqggyKkEIIcTIJoGBH85e1kNwSPKhEEKIEUwCAz+U6i0wkBYDIYQQI5cEBn70toCi dCUIIYQYySQw8MPZy8gDGZUghBBiJJPAwI9eehKkK0EIIcSIJoGBHw6n5BgIIYQYnSQw8EPJcEUh hBCjlAQGfsg8BkIIIUYrCQz86HUeA6fMYyCEEGLkksDAD5nHQAghxGilD3UFhqPeWgzMre389193 09RqpanZSmJsJI9++7pQV/mS8/HxbVxoOdc5AkRx1bTlTIibEupqCSHEqCeBgR+95RhY2m0UfHjE fdug1/HD7GXoddL4MhD7Txdyovao+3bqmBkSGAghxDAgVzM/eutK8GazOyirrAt1lS85TuXwuO1w 2kNdJSGEEEhg4JdzgAmGJSfPh7rKlxyH1wySdgkMhBBiWJDAwA/nAFoMAA6fqAl1lS85SnkFBg4J DIQQYjiQwMCPAcYFlEhgMGAO6UoQQohhSQIDPxwDXCiptrGV2obWUFf7kuK9UJUEBkIIMTxIYODH QFsMAEpOSqvBQPh0JUhgIIQQw4IEBn4MNPkQ4IgkIA6IdCUIIcTwJIGBHwNNPgQ4InkGA+KbfGgL dZWEEEIggYFfA5nHoEtZZR0dNseAnzdaOX26EmTfCSHEcCCBgR+D6UqwO5wcr7wQ6qpfMhxO6UoQ QojhSAIDPwbTlQBw5ITkGfSXb/KhdCUIIcRw0Gtg0NzQgB1oPnWQra9vp7w51NX9YqhBLq1ceLAy 1FW/ZDiU93BF6UoQQojhIGBgcOr1x4lJuJ6i6n08mJrByqwbmB7zFNWhrvEXoL8tBjqtxuP2wbJz nL0wSqKni+SUrgQhhBiWAgQGzRT87Enyij5gZuVbbMxYj0XVs45NFNeGusrB19/AYExcFHOmJnuU vb/neKirf0lQykmH1YjT4ToEZVSCEEIMDwECAztWMlk6J55jBU+Q+eB1GDtXaO6wjfxfdv1NPoyK COOmJTM8yt7bc3xQoxqGq3N1wWkBsVp1lH1+BSV7lnLm2EzOnh/caBAhhBBDK0BgEM+iW+pYHqFh yZOw9qsT2fWLa3mCFcyfoA91nYOuvy0GURFhXLdwGmEGnbvsXF0LB8vOhfotDIndB0/z7X/9K7/8 w3bKBjjiYl9pFds+LaO9wzeQdDidVJTMwmEPw+nQ03B+PB/sMPHtf/0rf3z9M05W14f6rQshxKgV 8Cq/Yt377FzyIa1jr+TmCQa2Jz9IYdW3SQl1jb8Aqp9LJUQZDURFhHH1/BQ++OyEu/zdwuNkzBgf 6rdxURxOJ38s2ItSsHPfSXbuO8miORP55k3z+3xv5tZ28v73QxrMFp75624yF03j5qtnMXPKGAD+ 9OY+WptifZ53rq6Fv7xzgL+8c4CU8XEsv2Iq116RytQJCaHeHUIIMWro/u3f/u3f3Lea97DCOJlP ImcyI+0yFi66gumT4gA9UxctYpJp5LcWAJypaWRHjwt9IDNTkrj28lQijAa2fVruLq+qNbPq+ssw 6HV9biNUlFL8/tVC4mMiSIiN9Ln/3d3H2PrJMY+y6tpm3i08zu6Dpwk36Jg8Lg6d1rfR6Xcvf8SR zrUjbHYHx05f4O2Pj7Lr8wrO1DSx+f2DfdavqcVK8fFzvLmrlO2flVPb0Ep4mJ4xcVFoNJo+ny+E EGJwPAOD8El84x/vwVH6Gtd9KYuf/99ZZs5KJXXyWMJH0YwHleebPFoAApmbNo7FcyczLtHE1t1H abO6EujsDifjE03M6PyFPBx9eriS37+6h7c+OsrxyjomJseS2BkgtHfY+fnz27C0+08IrDdb+PjA Kd7+uJTm1nbGjYnBFBkOuLof/vj6Z36f19hipbTCM3tVH9aOKaEemzUCpfxf8Jtb2zlyooZ3dh/j jV0lnKxuwGZ3kBgbRXjY6AhWhRDii+IZGAD6qATmLL6Zf/u3f+aeBTre+fV13HL3zynXpTJjRhpj TeGhrnPQnT7XwIdFJ/t83IL0CVyRPhGNRkNTi5VD5d0THJVVXuCWZbPR64ZnRPXbTR9RU98CwJma Jt7++CjHTl9gXGI0Hxad5JPiU31uw9ph51D5eV774DCHT5zHqWD9q4VY2vuboKpIvewQSZPOkD7L wTeuWUWHzcH5uuaAK1y2d9g5WVXPR/srePX9gyTFRzF9cmKod6cQQowYvVy1jCTPXMa61xQ1B95i 1tn3yJgYw9P7Rv44/f4mx0dFhLn/f8u16Rj03bvzQmMbr/ajyTwUSitq/SZI7jlUydpfv8mLbxR5 lH/9xnn85pFbWJoxhUCt+PtKq/nPl3ZSb7a4y7RaDf+4eilL5k5Gq/V9YvKUCqJimwDQ6GzcuHgG /+8fv8wr/343P8pexlWXTe41sHIqxQv/9ynNbe2h3qVCCDFiBGyHPfX646RmvUVh1QaemX8LGwHI 484FplDXOegGMlyxy9hEE7dffxl/7REM/PW9YlZeM8vdRD9cbH6/uPf33yMyio4I45tfnk90ZDhz p4/j7AUzr31whHd2H3N3nQTyrS/PJ+v6OWRdP4fGZgsfFJ1g26dlHD11gai4BpKnnHY/tucERzFR 4Xx56Uy+vHQmrZYOPjtyht0HT/PpoUpaLB0er2FubefF14v4p29eHerdKoQQI4JMcORHf4crRhrD PG5/6+bLiY02um9bO+w+v75D4cU3itzrOFTXmvlof4XH/Snj4wI+91s3u4KCLuPHxPD9O5fwlyfv 4kfZy0hPTfL7vNQJ8dx18+Xu23GmCFZdfxn/9WgW//3PNzNl1hGP1ge7w3/3Q9eQ0H++73o2P3U3 v3nkFlZcmebxmLc+KqWssi7Uu1kIIUaEAC0GPSY4WvcEmQ+WeE1wNLITvvo70U5UhMHrdhj33rKA /3rlE3fZu4XHWHX9HNImhaYf/PS5Rl7eup+Xt+7nsmnJRBgNHl0laZMS+H3u7Xx+tJpX3z/I3iNn 3PclxUeRdd1lfrdrDNO7f9VXVDfw3p7jbPu0jHqzBa1Ww4/vWR5wVMbYMVHowzwDAXs/pkTWabXM nT6OmSlJlFbUUF3r6tZyKsUzr3zCb354q98uCyGEEP0X4ArfPcERQEFV9wRHFaNggiNHf7sSvFoM AL5y7SwKPjzC6XONgCtfofDg6ZAFBls+Oer+/+ETNT7333nDPACumDWBK2ZNoKK6gb9tP8T2vWXc 99WFHpM3BZI6IZ5/uP0qvpO1iKKSKhqbre45C/xx+pkoYiBrJYQZdDy4eik/ffZdd9mRkzXc8ehG 0lOTmD01mfSUJGalJnm04AghhOhbwMyuFeveZ2fBZrYUVnDbBAO25AcprHqy/xMcWQ9yj+Ye9lnd Bbz+i3tYsWIFK1asYv78Fbxc6vrFV771aVZoNGg0Gh5/eQ9dl4jm8q2sme8qX/X4y1Tbey+nuZSn 16xAo9GgWfU4e2oHN31z/1sMfAMDnVbLLdeme5TVNLRe5Mc0ODa7g/f3lAW8Pyk+iusXTvMoS50Q z4+yl7HxF9/khiunD+j1dFotV1022WeaaG9OPyspDnQRpasum8zSjCkeZa2WDopKqtj49uf89Pfv svqxTXzn56/S0GwZ0LaFEGI0C5zyrddjq9/HU0tSWfPiJ9TXlFJp7scJ1l7L1hef5p7FGWykGVtX fpq9gs1PbOTeJ5/lhWd/wcYN+Vw/MQL7qVeZvvJhVm0roaJoM2/dvYQnd9WCvZwfT1/J+e9uo6Ki iDlP3s3K3+wKXI6VVx+czcPnV1FSVcbmOW+xJPk3DCYlYjDJhz2NTYj2uF3b0BKEj65vu4tP09Ri DXj/HSvmoguQ9R8fExG0Znl/LQZO5RzwWgnfv3NJn5NInalp4uk/fxyU9yGEECNRwEWUtj6ewA2/ rWNJNpxvj2LulRGsnn0/B619bbKNlvYYVj6yzrO45hjF5LJkWgz6iHHMXjCPCSY9JVufhZwtPLQi nZQFd7IhL4MnPjhBc8n7PM9ann1oBSkpC8gtyqP4sd1UBihvsB7h2Y1Q8OxDpE9I484nnieDTZQ2 DHynDGStBH+S4qM8btd2thjUNrQO+aJEr7xXzEf7K/wGM29/XOpxe0JSjHup6KiIMFZePWtI69Jf zgBzTtudA1thcVyiiR/fs4wr50zCFBkW8HEfHzjFjs/KB7BlIYQYvQIkDDRz7K0Mtnz0HDNePs7X mgyk37yO/IyFHD1vZ15KL3kG+hTuXHMf2A/yP/fvcBc3VJZSzJPMTn4LKAZyKKz/bwztkJHWvXTx 5KV3wg/20LIKYCpdgyNNkxcBv6SsdZXf8mPNNwKZjO26wziNOylm98kGlsXHD2inqH62GLz1RoHf cf1Wm+fzq2ua2Lx5MydqbOw6ZmX5LCNTkwz9eo3eOJ2Klz5pocOuMBk1XDYxnFnjDYTpNTRbnewr 9WypWJriIGx6FIcqOzAaNLz5xmsXXYfBaFNmv+V/+/vf0A0isfXyZJifFE5jm4Eas4PaZgen6+y0 tnd/Dr/Z+CGVR4uICJPkRCHE6LB69epBPU8fuLiY4uO1pIbHgDES7CVsL4YfxfTzxG33+vUXu5Qt O8u4cVkaemp5blUyOS+tYSNQTPdjjVHhUGzFjhGwuvMNMEaRSdeQtEDlPYes6YnNgKaB/QgFoD9x QUS4ga9/PfBOf+XTF+mwufrSbQ7FV27N4v92HKbjyD7eP2zh1mUpPHDHYsIMg0/m3FdaRcfOrQA0 WxWF5VYOnHHwpSXTO/fQcfdjZ6WMYc39WYN+raFU3Xia4q0f+pTf+tVbiDbGDMlrnD7XyPeffA2b 3fUZtNsUB2ujufnqmUweF8fk5FgijBcfnAkhxEgT4Kpk4lvP55O8MJnHADLMaB7eAWsL+MvAfnx3 v1DMBGaPTet8wSSuW5nJA002bEagsXvSmoqPtkLuT4mzHQLa3QGAtWIPO7iTP4Tht3xaZ/J5q/uO k7xWDD+dNvAKO/uxvKL3UEVvSfFRVNV0/zKuqW/hTE2T+/abu0o5XH6ef/nuCqaMixvUPvU3bbGl 3cbrH5b4lH/lmvT+bPIL4XT6378DTUDszZRxcdx76wL+8Nped9mB42c5cPys+3ZCbCSTk2OZmBzj +kuKZUJSDOPHmGQNBiFESDntVlrLCrDVl6INM6EJi0ZrMKENi0ZriEYTZkJriEYbFo3GEI02LAqN dmh+7AQ8+yUtfghzxbW8s6OYOsKYMfsFli9OG/QMBsc3TmfhY3lUWB5louUgzz2wg+xNY5iRvgoW /iu7fvAuy0xl/PHhHeQUbMA0I45MFrJh13d5dJmJN//4MOQUMHnGJL/l8aapfCOjmEde2EXRo8so e/MldrCWDUkDr2t/ll0OlF/g3n/x0R6BQW1Dq3sIY5eT1Q38IK+AnK9dxVeuSR9wst/uA6f79Thj uN5n9EEoBc4xGLrAAODOG+by8f4Kn4WbutQ3tVHf1OYRLHRJiI3k6zfO42sr5oZ6dwkhRgjr2U9p O/k2Gl14jwu7qfNCH915Owpb40lajr6Cs71xQNvX6MLRGKJc2zNEEX/Vo2jDBj5bccDrfPmeXTBn GXfet8BdVrprDzFLF9P/qQxi6IpfFjxURv7B6aRGPAZAxtpNfHBXGiZy2Ja3neUTXY/MyNnE+7el ACls2JZH6vKJrlYL1lJUfxt6E/7LgW+/voUtqcsxuJo52FD0fv+HV/bQn+RDf3MY9JTslYBY0+DZ YtDF2mHn6b98wpaPj/KPX1/KnGlj+1XH46cvUNvYPQwyzKAjOiLMY62CLtcvnDasms2dyuG3fKgD A51Wy4/vWe7RpdBf9U1tPP/3T5k6MYErZk0I2b4SQgxfzo5mWo5uxmlrcV/k3Rf4Hhd7ZbfQXLKR jtrii3/RXihHO8rRjtNa7yrQDu6nvM+zrA3VVNVX8tSS5bRtKOSXy8ZgsYGB4zy4fCU/r1dM6E/r vHEB29VrPW6n8dBLigdfsGJBj8nY9dJGVjz6GraHmrFgwGTsnpAmZcWjKNtDNFvAZOq73JhyM68p G83NFgwmE4Od2qY/wxUj+9GV0NOREzVYe1l18HhlHWt//SbLr5jK3V+5nKkTEnrdvnc3wpVzJvEv 313BJwdO8c7uY3xWcsY9w+Fw6kaAL6YrocuUcXE889htHD9dx+lzjZw+10jl+UbOXmju83N2KsWv /riDZ/95lc/nKYQYmWyNZXTUlXQ2z3de4Dt/yWsNUWh0rini207voPnIn1Adw3RhQY0WrT5iUE/1 CQyOFzxGxv2uJZO4f0nn4kkuGTnrSR9kjoH7BY1GTH7LTX7L0RsxmQZQjh6T6eIWevJu6tZofFdc 7LvFwHMug31Hqz1ux0UbsXbYsXZ4Xgx3fn6SnZ+f5NrLU1l1/RwyZoz3u/2PD3gGBlfPT0Gv07J8 wVSWL5hKbUMr7+w+xrHTFwKuZxAqAbsSHEMfGABMnZDgE2g5HE7OXmjmTE0TVTVmqmqbqK41U11r 5lxd92iOphYrv/ifbTz5g5v77D4SQgwPSik6ag+AVt+jHz6q1wulcthoOfoKrWWvA730J2v1aHRG lO0LmJ9GF0bUtFvRmybj7GhB2ZpxdrTgtLXg7GhG2Vo6b7e66tPj3KoxDP7HjE9gMO++l1D3/Rf/ smQulkd38NTXZozwlRF8ef+QNOh17hEGXfrOMfD8UOqb2jxuL5oziftvW8QL//cpHxSd8Hn+R/sr +Gh/BROTY7h56UyuXziNsYmugOfsBTMV1d0TNGg1GhbPneLz+tlfuSLUuzLA/vXfrB+MFoNAdDot k8bGMmlsrM99G9/+nD+9tc99u7Siltt//BKRRgMJMZEkxEaQEBtJYmwkd918OaYei0wJIULLYW2g 6fP/8t9sr9F19sF39cNHuvvjO+oO4Wip7vsFnHaUc/BBQdiYeUSmfgmn3YLqcZF3XfhbO/9tIWzM PKLTv4Euov/T6XdvsxXl7Oj387wFuOYbueFrmdxwx0x+SwbrNuRxb9b1pMSPjnnnvecxGExgkOw1 +6G3yWNjSYqP4vHvZHLrsnTW/22P3xUCq2rM/KHgM/5Q8BnpqUksXzCV+ibPPIJ508cRE3XpXJwC tRh8kYFBb+5eeTmlFbV8erjSo7zNaqPN2uSRK1J2uo4n/+lm9DrtQF9GCDEIymmntaygs18/Co0+ Cq0hEo0hCmVrwXz4fwM37ysHqsOMo8PMwLKOetH5q14bZnJd4G0tPS74ra5f+bZWtIZoTJfdi3H8 4qDtG60+AvQR6Li4VuKAgcGKR19CPfoCpw7uZetrz5OasJKM7HX84gd3cdPitEH3318KvKfmNeh9 T/p9thjE9d6MM7nHEMWMGeN59p9XsfvgaTa+tY/jAZYQLq2o9Zthf/X8KVxKvqhRCYOl0Wj45/uu 48F/L+hzpsoDx8+S/+eP+VH2slBXW4hLilIKlAPNABLkHJYLNH72a2wNx/v9nMHQGhMJG3NZ9wXe 1vlr3tYKPc5TYclXEJPxD+gjky/i1YafPj4RIzGJSYxPngxAcVURv1vyBFmsxeH4NVrtyPyV5PDT YuAtso8s/wijgeiIMFos/ptzJiX7NmEvnTeFpfOmsK+0irc/PsonB05hd/Q9dvLqjMGMvQgdf4so wfAJDACiI8N54h9u4L9e+cQ1rNHchs3u/7N4Z/cxxiVG89Vls4mKDEM3Qr8XQgwVZW+n8fOnaT+7 B3RhaPWRrmZ9fRQaQwQafWRnWQQafQRafSRoNLQc+9uA+vZ1URPQaHU4ba04bS3g6Lt5PSLlRkxz 7kVriAxYd6etBeXoQB89vs/tXYoCBAZ29r36G368+jF2AFm56yksq2FxWhLQwFOaBA4f/g7z5s0L df2DwrfFwDcw6E8iWlJ8lN/AQKvRMCEp8Ax/C9InsiB9IuYWKzs+O8EH+05wuPy838emTUpw5x5c KgJ2JQQp+XCwpk9OJP/HX3XfNrdYqTO3UdfUxn/95RPOXuhuTfjfN/fxv2+68hKS4qJ46uGvMDF5 aGZxFOJSoZy2PifZcVjqaNjz79jNJzsLOnA6OqC9ceia9zU6otO/SdT0LDSa7kBdOW2dzfutHq0A Xf8a4mcSntT7dU2jD0env3S6bgcjQGBgofS9g6wqKOSVmxaT5NFvEM99ZcUkpMwOdd2DxnsegzCD n8DA2J/AIJqT1b6rOI1NjPa7TW8x0Uayrp9D1vVzqG1o5aP9FXx6qJLisrPuX69LL7HWAhj+XQmB xEQbiYk2MnVCAr988CYe/o83/AZ+tY2t5D6zld/96FYSYiMH8UpCXHraTm/DvP85ADR6Ixp916/9 iM7brrKO2uIBT9zjT1jy5Rhip3lc2JW9DdAQM/c+DHG+y8ZrtAZ0xjgwxoV6dw1rAadEvuu5l8Ba S+mRPZS2QtKUmaSnuMYqJqeNzJaCLs5+dCX0p8UgOcF/noG/TPi+JMVHcXvmZdyeeRmWdhv7j1bz 6eFKll8xNdS7a8ACdSUMl+TD/pg8No4ncm4g97+2+nQ9AZyra+Zfnn2H/1x7iwxzFCNey/H/o6Vk k/u2srd1XqQZulaALhot0el3dbYGyKJowRAwx6D54Mtcm3E3PQd8ZOXv5NWHlo344YveXQlhfpMP +55JMCne/8iEKWPjLqp+EeEGlmakXJKtBTB0yy6H2uUzJ/DD7GW8tuMwzW0dmFuttFm730P5mXr+ Ma+AScmxGMP1GMP0hBv0GMP1hBl0GMP0hBn0nf/qCO+8PzxMR5hBT9qkBL9BqRBfBOW0Yz74B5zW BtdUu7pwNPow9//RhaPRhmE3n8RyevugXiN83JXEzn8A5bSj7G04bZbOf1tRdgvKbsFpa3P/X9kt RE69mbDEOaHePSNawGWXX86+m6nrd7JnzTKMQEPpq1w/+18pun87iy+tLu0B8zePgbd+tRjED12L wUgSeLjikP+2CLovLZ7BlxbPcNXf4eTfnn+fPYe6hzl2TZo0GAtnT+Tfcm6UBZ3EF045bTTu/TXt 5z8L2mtETV9F9Oy7e/zq7/94fRFcAZMPm8jiie8ucw9LjE+/le9mPNu9euEI1p+uhMh+5Rj4Dwwm j/rAIMCohGGWfDhQOp2Wn353BY89vYUjJ2suentFJVU8/t/v8Mvv3zSs1roQI9vFBAUafSRxVz2K ISYFZbfitFtdv/Qdnf/arSi7FW1EIsZxi0L9VkUAAQKDCNKXFHD/k1v5IPdm4vV2Tu16noeLZ1A2 wlsLYChHJfjvSph8kV0Jl7qR0pXgT3iYnl88eBOP/PpNn9U0B+Ng2Tke+68tfHnpTHRaDTqdFp1W 2+P/GtdtnRadzvV/R0MplvICNBonOhyEj5lF3GX3kBgbKfkOo5hSTpqKfou1ejdo9K75A7Q9/tXo QKt3LcJjuTDg7WvD44lf+lMMMZ1dnGEmpCPs0uQVGDTz3IoYHtjRdXslCU/0vD+DCxZIG+HBge88 Bp45BmEGXb9muhsT55uRHhURRnzM4Ba2GCkCL6J06XUl+GOKDOe3P7qVM+ebsHbYae9cE6Pnv+4/ m6PzXzsdNgftNjuV55o4X989VjvQxFa981ou+u9/IyE2knVrbmRmyvBaO0N8MZoP/dEVFAAoO8ph Bwf0tWScLnoC0TNXoxwd7tX7lKMdHB2dZR0oZSc6/ZsjbqKf0corMDDx9Y0V3Bzwh5uBsSM8KAA/ yYdeQwv7+6vLoNeREBPhsRTyaO9GgJHdYtDFFBnO7KmDO0m2tLXz+H+/M4hgoHf1TW388Ldv8ei9 17F8waU3mkUMXsuxv9F2cuuAn6eLnkDC1T9HZ7zI1fPEJcWnKyF+Qgqj/RDwzjGIMBoYlxiN3eHE 7nAGTCr0Jyk+SgIDL8NhEaXhLDoynH//p5v56bPvcijAxFaD1WFz8Ms/bOfyXePR+xltc9MS14Jd Yviyt1TR9PkzKOVEQ8/hen6G7mlcXQj2xvIBv44EBaOXpDv74T3BUcq4eH7w9asHta3khGiOnuru r5s0yvMLoJeuhEs8+XAoRRrD+NUPbublLZ/TarHhcDpxOBUOh9Pz/47O/zudOByuf22WBmwtNTiV Bica7CqM6hbP1U32Hzvr93U/O1LF0YpavrfqSnSyMNSw47RbaPj0KRwtVYPehsYQ5brgR43tXCnQ 3v2vcrhv66LGog0bBU3EwocEBn54dyVotYOfROOfvnE193910aBaG0aqS3Xmwy+aMUzPd7KuHPDz Wk+8RfOhAvdtbcQYPo/+Ifl//rhfa2/8bfshjp2+wDXzU1ytZxr48tKZsrz0MND0+TMXFRSgDSP+ qlwMsamhfitiGJPAwA/vFgPtRcyuFWeKIM40upMNvQUcriiBwdDwTuJ0dPDlpTOZkBTDz59/H3Nr e5+bOFh2joNl59y3X/vgCP/yncxB502IwVNOO8ppo+3E265FhwZNS9yiRwhLTA/1WxLDnAQGfnjn GFxMi4HwFXiCIwkMhoLyCrxU54py86aP4/e5t1PhZ/2ONmsHz//9U2obW/1us6a+hR/+5k2+t+oq 7rhhLuLiNR95CZv5lOvz6crud3Zn+uO0oRw2wP/3RR8zlZj5OT1K/Iwv6FGkNUSiN00K9dsWlwAJ DPzw7kqQ+biHVsCuBMkxGBpeAZbqsdRsUnxUwIm3Lp81gf/3xx3sP1rt936HU/Hc3/dw4PhZfpy9 jJhoI2JwLGd20VpWMOjnawzRxF31ExkeKIJCAgM/fFoMJC4YUiNhEaXhTPl01ThRTrtrIptexEYb efIHX+bVbYe40NCKRqOhqcXCjs9OeDyu8OBp1vzq/1h71zUkxkZhszuINBpIGS/Z6/3hsDZgPviH i9iChriFj0hQIIJGAgM/fHIMtJKdPZS8928XyTEYIn5yOJSjvc/AAECn1fKNL2V4lF23YBr/+dJO jyWm65ra+Nnv3/N43B03zOX+ry7q15Lio5n5wHqUraX/T9Bo0WjDQGdAozUQNe0WwpPnh/ptiBFM AgM/vK9b0pMwtGQeg+BSTn+BQQcYBjci5ur5Kfx+0ir+3x939Drp0t+2HWJfSRX/fN/1TJ2YEOrd MGw4O1pwdjThbDfTceEQ7eeLPO6PmvE1wsctQqPtWrkwDI3WALowNDqDa6piIb5AEhj44duVIJHB UArUlSAtBkPEz37smWcwGGMTTfz2h7fyv28W8cp7xQRo9OFkdQM/eKqAb9+6kDtvmDeqEnfbaw5g rdqFo70Jp7URZ3sjzg6z3xacLoaEWUSnfxONRlolxfAhgYEfziGcx0D4CtSVIBMcDQ2lArQYXCSd Tst3sq7kilkTePbVQhwORZhBR725jcZmq/txNruT/3ltL7uLT/GTe69jQlJMqHdJ0DntVpr2/zdO a33/n6QLI/byH0hQIIYdCQz8kFEJwSXzGASZv/3r7Hvugv66In0iL/z0DvftVksHz/x1N9s+LfN4 3OETNTzwq//jO1mLyLpuzoj+HrUe//vAggLANDsbffT4UFddCB8SGPjhvbqiTloMhpTkGASZMzgt BoFERYTx2LevY+m8KeT/5WOae0ygZO2w8+zmQnbuO8mPspczMXnktR7YW8/TWv5GwPs1OiPa8Fi0 4TFow1z/6iLHEjl1ZairLoRfEhj4IS0GwSWjEoIrWF0JfVm+YCpz08byu5c/ovBQpcd9h8rPs+ZX f+feWxZwxw1z0Y2gkT7NR/4EPVYG1YbHEXfVo+jC49GGx6LR9W81ViGGi5Hz7RxCknwYXDKPQZAF IfmwvxJiI1n3/Zv4UfYyIo0Gj/s6bA7+57W9PPTU65RV1oV6Lw2aUgpHWy3tNftpOfpXn2mKTXPu ISx+JrrIJAkKxCVJWgz8kOTD4Ao886FtgFsS/qgA8xh8kb68dCYLZ0/kdy9/zKeHPVsPjlfW8YOn CrhjxVzuuWUBxrDhexpytjdhM5/Cbj6N3XwKe3Ml9uYzKIfV7+MN8TMwTloe6moLcVGG7zcyhKQr IbhkrYQg+4JzDAIZExfFLx+8iff3HOf3f9vjkXvgdCo2v3+QnftO8k/fuJqr5k4O6S4D14yEtsZy bI1l2JtOYms6OeCEwpi535XzhbjkSWDghyyiFFyB5zFwDHBLwp/h0GLQ042LZ7BwziSe/etuPtx3 0uO+8/Ut/PT373LN/BQeXL3UvY6DtcMe1JYE5bRjazqBrf4otoZjdNQfw2m9uO6NiMnXY4ifHvwd KkSQSWDgx1Auuyx8BWoxcCoJDIZECHMMAok3RfAv313BDVed5r9e+YTaBs9VHD8+cIrPjpwhKiKM phYrDqfihivT+N7tV5EYG3nRr6+cdmwNZXRcOEhH3WE6Go7BRewTjT4SvWkyetNE9NET0UVPJCxx Tkj3sRBDRQIDP2RK5OAKFBiAK89ArzMMYGvCh7/96wxtYNBlybwpzJ85nj+9uY//23HYIwhvtzlo t1nct7ftLefj4lPcvfIKvpZ5GQb9wKYGtrecpb1mPx01n9NRd3hwrSZaPXrTFAwxKehjpqCPSUFv moTOKFM+i5FLAgM/nE7PE6ssojS0emsZsDvtEhhcJKX8tRiErivBW0S4gTV3LOaGq9J4+i+f9Lr+ grXdzh9e28uWj4+y5mtXsTQjpZf37cRWfxTrub20n9uLo/XswCqmNWCITcUQl4YhNg193DT00ZPQ aGWtAjG6SGDgh1eKgSy7PMR6bTGQBMSLN0ySD/syffIY8n/8Vd7++Ch/LNhLc1vgOlbXmvnX597n ilkTeOCOxe5FmpRS2OpLsFZ/grW6EGd7Y79fXxseT1jibAwJswiLn4U+NrVfK1AKMdLJt8AP7xYD yTIeWt77tycZmXDx1CUSGIDru3XLteksuzyVmoYWYqKNmCLDeX9PGS++8ZlPsPD50WoeePL/+NKi SXx97gWMDbtwWi7067W0xgTCxsx1/SVehj5qbKjfvhDDkgQGfni3GMiUyEOr9xwDCQwu2jAbldAf MdFGYqKN7ttfXT6b6xZM5X/f3MdbH5V65CIoBe/uPcOH++zcMjWOm1PqMer9HFMaPWFj5hCefAXh yZejN4V+SKQQlwIJDPyQeQyCq7ccA2kxuHj+cwyGZ4tBb2KijfzTN6/m1uXpPPe3Pewrrfa4v92h 5+9lM9hWOYXb045z3cQz6MOMhI9dhHH8lYQlX4FWHxHqtyHEJUcCAz+85zGQwGBoSVdCkDkvvRaD 3kydkMC//9NKCg+e5vnNOzlT5/lemtrDefHIXN45t4D7b1vCdQumy3dWiIsg6fZ+eM9jIF0JQ6v3 5EOZFvlihWoRpWBbMm8K/5nVyL2zD2EK830/Z+s7+NWLO/nHvAKfaZiFEP0ngYEf0pUQXL0OV3TI JEcXzV+LwTCZx+BiKOXAVvMZN06p5D+XfcBt08oI1/t+N8sq6/jps+/yyK/foPj4AIcsCiEkMPBH pkQOrt5aDKQr4eL5azG4mFn+houOC0dQthYAIvQOVs85y4afr+aWa2f5/Y4ePlHDj3/3No/mv82h 8vPYHU7OXmim+PhZzK2XbteKEMEmOQZ+yJTIwSVdCUHmHN4THA1W+znP5Y2N464kNs7Ew9+6ljtv mMeLbxT5rMUAsP/YWfb/5k2PsohwA7ddN5s7VswlziQJikL0JC0GfkhXQnA5nb2NSpCuhIs2AnMM lFJYz37qURY+frH7/xOTY/mX767g2X9exeJ+rNRoabfxyrvF3POzV3jub3uoa2zt8zlCjBbSYuCH dCUEl7QYBNdITD60NZZ5LIGs0RkJT8rwedz0yYn84vs3UXKyhhffKOLzo9W9brfd5uBv2w/x+s4j 3LRkBl//0nzGjzGF+u0KEVISGPjh8A4MJC4YUr3mGMgERxfvEpr5sD+UcmKpeM+jLHzsFWh0YQGf M3tqMnkPreRg2Tn+9OY+Dhw/i0YD8TGR2OwOmr1yDGx2J299dJQtHx9j+cKp3H79ZQCcq2umua2d 5VdMlS4HMWpIYOCHT1eCRAZDqreuBLt0JVwU17Gr/NxhRykHGs2ltSBQR8MxzMUvYG/yzB0IH3dV v54/b/o4/mPtV6hraiM22ohep6W9w87bH5fyynsHqW9q83i8Uyk++OwEH3x2wqP8+b9/ypcWz+DO G+YyMTk21LtFiKCSwMAPnymRJcdgSPU+KkG6Ei5KL6M6lKMDzSUyE6BytNNc8jJtJ97yvVOjJ3zs wgFtLzE20v3/8DA9t2fO5ZZrZ/Nu4XH++t4BztW19Pr8DpuDtz4q5e2PS1kybwp3rJhLxozxod5N QgSFBAZ+SPJhcMnqisGjepkjAkcHXAKBQUf9UZo+f8bvsskanRHTnHvQGiIHsWVPYQYdty5LZ+XV M9lRdIK/vltMxdmGXp+jFOwuPs3u4tNMn5zI1zIv47qF0zDoL62WGCF6I4GBH5J8GFwOWSsheHrZ t8M9z0ApRevxv9FS+lfAN3g0TrwW02XfRmeMH9LX1em03HjVdG64Mo09hyp55d0DVJxtZFxiNGMT TJSdqaOm3rdFoayyjqf+tJMXXtvLrdemc8u16STEXnzAIkSoSWDgh8xjEFxKVlcMnj66EoYrZ0cz jfvy6ajZ73OfLjKZmPnfJzxpXlDroNFoWDJvCkvmTfEodzic7Pz8JK++f5DjlXU+z2swW3jp7c/5 8zsHWL5gKlnXzWH21OSQ7k8hLoYEBn4opyQfBpMsohQ8vXUlKOfwnOTI1nSShj3/jtPqe9GNSL0J 05x70eqNg9jy0NDptGQuSiNzURrFx8/y9+2H2X3wFF6/H7A7nGzfW872veVMn5zIbctnk7kojfAw Oc2KS4scsX5Ii0FwKX9Z850kx+Ai9TKqQ9mHX4uB9XwRTZ/9FuWwepRrwmKIW/AQ4cmXh7qKHjJm jCdjxniqa80UfHiEd3Yfo83qmzBbVlnHbzZ9xPN//5QbF8/g1mXpTBkXF+rqC9EvEhj4IYFB8PQ1 s6G0GFyc3lsMhldg0HpyC80HN+CdT2BISCdu0Q/RGRNCXcWAJiTF8P07l/DtWxfwbuFxXv+whDM1 TT6Pa7F08NoHh3ntg8PMmz6OW5elc838VMIMkqwohi8JDPxwSldC0PS2siJIi8FF663FYBitl2A5 s4vmg3/wKY+cdgumOfei0V4aF85IYxirrr+MrOvmUFRaxesfHGHP4UqfbgaAg2XnOFh2DlNUOF9a PJ2VV88iZfzQJlIKMRQkMPBDSYtB0DiV6vV+ST68OEpdGsmHHRcOepVoiZn3HSKn3hzqqg2KRqNh 0exJLJo9ifN1zbz5USlbPzlGU4vV57HNre38ffth/r79MLOnJnPz1TO5fsE0IoyGUL8NIQAJDPyS 4YrB45SuhOC6RIYr2hqOe9yOvfz7REzJDHW1hsTYRBPfzbqSe29ZwEf7K3hzVykHy875fWzJyRpK Ttbw+82FLF8wlZuWzGDe9HEyd4oIKQkM/JAcg+DpqytBAoOLo3oLvIZJV4LTbsHefMajLHx8/6Y4 vpQY9Dr3aIbT5xp5++NS3ttT5rNOA4C1w867hcd5t/A44xJNLLsilRZLB2fON9HYbGFpRgo3XjWd 1AnS9SCCTwIDP7xbuyUuGDp9diVIYHBxnMO/xcDeeIKe6znooieiNUSFulpBNWVcHA/csYTv3HYl Hx+oYMsnx9gfYOXHc3XNbH7fs6ul8r1i/vpeMdMnJ3LDldO5ftE0j2mehRhKEhh48e5GAJkSeShJ i0Fw9ZpjMExGJXR4dSMY4meEukpfmDBDdyvC2Qtm3tl9nPcKj1Pb2Nqv55dV1lFWWccL//cpl88a z4orp3PN/BSiIsL69Xwh+kMCAy8+iYeSXzCkepvcCMDukEWULkovs0oOl1EJtkbPwCAsbnqoqxQS 48fEcN9XF3LvLQvYV1rFu4XH+fjAKWz2vlcYdSrFvtJq9pVWk/9nHYvnTiZz0TQWz51MmEFO6+Li yBHkRfILgqu3BZRAll2+aJfAlMi2Udxi4I9Wq2HRnEksmjOJlrZ2Pig6yYmqOiYmxTJ5bCzmtna2 7Snj86PVfrvibHYHH+2v4KP9FUSEG7g6YwrXLZzGwtkTZXEnMSgSGHjxmcNAAoMhJV0JwdXrBEfD oMXAYW3Aaa3vLtAa0MekhLpaw0Z0ZDi3Lkv3Kb/xqunUNbWx47Nytn1aTvmZOr/Pt7Tb2La3nG17 y4k0GrhmfgrLF0xlQboECaL/JDDw4h2R66QrYUj13WIgXQkXZZgnH/q0FsRORaOV01B/JMZGcucN 87jzhnmcPtfI9r3l7PisnLMXmv0+vs1q4709Zby3p4xIo4GlGSksuzyVRXMmSneD6JUcHV68cwyk xWBo9ZVjIC0GF0cN83kMvPMLDKM0v+BiTRkXx31fXch9X11IaUUtOz4r58N9J6lvavP7+DarjW2f lrHt0zKM4Xquumwy116eylWXTSLSKImLwpMEBl5kcqPg8u5K0OsMHgmHfa2lIPrQa45B8LsSHNZ6 7M1nCE/K8Hu/5BcMvfTUJNJTk1jztcUcKj/Hh0Un2bX/JI3NVr+Pt7bb2bnvJDv3ncSg13LFrIlc Mz+FJRlTiDdFhPrtiGFAAgMv3qMVJS4YWt5dCWG6cI/AQEYlXJzeWgz4AloMWo5uRtla/AYGSils jeUeZRIYDB2tVuNe/fHBry/hYNk5du47yUf7KwIGCTa7k08PV/Lp4Uo0f4Y5U5NZmpHC0owpTB4b F+q3JEJEAgMv3k3d0pUwtHwCA30YbT2uVzLB0UVyhm51RXvLWSynt4FGh9PWhtbgOQGPo6UKZbe4 b2sM0eijxoV2f41QOq2Wy2dO4PKZE/jB16/mUPk5dn3uGrlQF6C7QSk4fKKGwydq+J/X9jIxOYal 81KYMy2Z2oYWKs420tzazoL0CVw1dzLJ8dGhfpsiSCQw8OI9Gki6EoaWd+Bl0Hn2b0pgcJFCmGPQ Uvpn1zwKyon17G4ip9zgcb+l6iOP29Ja8MXwaElYvYTSilo+3l/BRwcqqK5tDvi8qhozr247CNs8 yz/aXwHAtIkJXHXZJK6aO4XZU5PQabWhfqtiiEhg4EXmMQgu7xyDMF24x21JPrw4oRquaGs6gbX6 E/dta+VOj8DA1lBG6/G/ezwnTAKDL5xGo2H21GRmT03me7dfxcnqej45cIpPDpzieGXdgLZ1oqqe E1X1/OXdYqIjwlg4eyJXXjaZRbMnkiDTNV/SJDDwIvMYBJd3V4JB79liIIHBRQrRBEfNRzZ53O6o O4zDcgFdxBiUvZ3GffkeszJqw2OJTP1yqPfWqDd1QgJTJyRw98orqG1oZffBU+w+cJoDx89idzj7 vZ0WSwcf7jvJh/tOApA2KZFFcyZy5ZxJzJk2Fr1OWhMuJRIYeJEpkYPLJzDQeQcGMirhYqgQzGPQ fuEQHbUHfMotZ3YRPeN2zIdfxNF61uO+mMsfRBseG9qdJTwkxUdx2/I53LZ8Dm3WDj47UkXhwdNU 1ZqZPDaW1PHx2B1O9hyupORETa8LopWfqaP8TB2vvFuMMVzP5TPGs3D2JBbMniBJjZcACQy8SFdC cDm9Llw6rQ6dVucRENgddvQ6OTQHJQQ5Bi0lm/yWW8/sxBCXhuXUex7lESk3YRy7MHT7SPQp0hjG 8gVTWb5gqs993/zyfMyt7RSVnOHTQ5XsPXIGc2vgbipru53CQ5UUHqoEXAHIwvSJXJE+gStmTSBO hkgOO3L29SLzGASXd4uBVqNFp9V7BAYOpwQGg9XrcEVnB0qpIe0es5791Gdugi725kqa9uV7lOmi JhBz2bdDvZvERYqJCnevEul0Ko6eqmXvkTN8duQMR0/V0tvq6rUNrWzdfYytu48BriTGK2a5goR5 08cRYTSE+u2NenL29SIzHwZXoMAAun9x2J02wjGGuqqXpr5yNJwd4JXwOVhKKddIhN5err2pxy0N sQt+gEY/NK8vhgettjuh8d5bFmBusVJUWkVRSRWflVQFnI2xS1cS49+2H0Kr1TArJYnLZ47n8pnj mTNtLOFhcpn6oske9+JwSldCMHmPStBqdOi95sqXPIPBU30sUqUcHWiGKDDoqPkce3NljxINUTPv pPXYZr+Pj0i5gbD4maHeRSLIYqKN7tYEgJNV9RSVVFFUWsXBsnN02AIfo06nouRkDSUna/jzOwfQ 67TMnppMxoxxzJ85njlTk2Wdhy/AF5YqWvrq48yfv4JVq1axatUq7lm1isdfPghYef0X97BixQpW rFjF/PkreLnUNba2uXwra+Zr0Gg0rHr8Zao7fwwFKqe5lKfXrECj0aBZ9Th7agee4S7Jh8HlPY+B Vqv1CQxk9sOL4Ow7MBgqrSfe8rhtnHgt0TNuR6P37TPWGKIxzb471HtHhMDUiQnceeM8nvzBzfz9 P+4h76GVfPOmDGZMGUNfv7vsDicHy86xact+Hs3fwu0/fokf/uZNXnyjiKKSKiztcq4Ihi8s9Jp4 5bd45pmVQBhRYZUsXLKanO/kg72CzU9s5N7CEpbF2mhrg8SJEWAv58fTV3I+fxsVWXE8l7qQlXGT OfDDCf7LH72SVx+czcPN+ZRUvcChZ77GkuQ4atSjJA2gnpJ8GFyBuxK6ySRHg9efFoOhYG8+4zMS ISrtFjS6MIzjl2Kp3O5xn2nO3WjDTKHePSLEwgw6dz7Bd7Kgua2dA8fOsv9oNZ8frabyfFOvz7fZ nRwqP8+h8vOA64fbjMmJzJs+nrlpY5mbNhZjuIHT5xqpONtAuEHH3LRxxMdIguNAfGGBgSllHss6 l10/+NyTsLaA/74tBapfp5hc/mVaDPo2A7PTk9ADzQff53nWUvXQCiYAuUV5PLlwN5UrY/2WNzwU xbMboaDqIdInQPoTz5PxZA6lDY+SFN//evrOY/BF7aHRwW9Xgk66EoZMny0GQzPJUeuJtz1uGxJm uVdKNE5e7hEYGOKmEzHlxlDvGTEMmSLDufbyVK69PBWAuqY29h+tdgULx85yrq651+e7Eh8vcPTU Bdcsjbh+zHn/wJuQZOKyaWNdf2ljmTIuTvLHevGFd9bYT71OxgMn2Vl/G3qgobKUYp5kdvJbQDGQ Q2H9fzOHdmAqXb8xTJMXAb+krHWV3/JjzTcCmYztusM4jTspZvfJBpbF9z8y8O1KkIk5hlL/Wgyk eXDQVO+tLUPRYuDsaMFy5gOPsqhpt7j/H5Z4GVpjIk5rHaAhJuMf5CQs+iUxNpIbrprODVe5gsya hhYOHDtL8fFzHOhHoAD4nV+huraZ6tpm3ttTBkB0RBizpyYzZ1oyc6YmMys1SZaf7uELDgzsvP+r LMgrZFnXtTp2KVt2lnHjsjT01PLcqmRyXlrDR5kAVtynOWMUmXRN2RmovOeUnnpiM6BpgNcY79UV mxob2Lx588A2IgI67zzlcbuiogKravUo2/re28Rpk0Nd1UvSTP1JxuoC3//Bjvdoch7o/wb9mKgr ZZq+O8CwOiN446NK0HR/T1J1SUzW11Ftn8au9/YB+0K9a8QlbEo4TJmnocUazdlGO2ebHJxrdNBk 6f/sjD21WDrYe+QMe4+c6SxRJETpGBurI8mkIzlGR1yk9pIPaFevXj2o532xgUFzEU89D5vK5ndX IGYCs8emdVYkietWZvJAkw1sAO3uAMBasYcd3MkfwvyXT+sc3dbqvuMkrxXDT6cNoB8B366ExMQE Vq/O+kJ300j28fH3OVlU7L49PW06lo42Pj/dHdRdcdV8Fk29dlDbb2qrJzYyIdRvM2QaP6vEWn0q 4P3Lr11yUZMLKaWoff/7OLsXSWTMZV9j9YzbPR5nMy+h/pMnuPzmn7EgTFbhE8HR0GzhUNk5Dpad 51D5OU6cqcepFGMTokmdEE9Ti5Xjpy/4jDbzpaG+1Ul9q5MS18WHqIgw0lOSSJ+aRHpqErNSkkbN ZExfaGDQfHwPO8hlY1r3GPXjG6ez8LE8KiyPMtFykOce2EH2pjGYZlxLJgvZsOu7PLrMxJt/fBhy Cpg8Y5Lf8njTVL6RUcwjL+yi6NFllL35EjtYy4aBZB4i8xgEm29Xgg5ThOfUuM3WpoFs0sMb+/9C 9tUPhvpthoxv8qEG6HFMX2RXgq2+BKflQneBNozIFN/8AUPMZOKv/AlaCQpEEMWbIlh2xVSWXeGa obHN6jq+e3YLtHfYOXqqlsPl5zl04jwlJ2posfT9PWi1dLjmYyitcpeNS4xmVoorSJiVmsT0yYlE hI+8CZm+0MCgqngvZF5Lz9zkBQ+VkX9wOqkRjwGQsXYTH9zlGv+6YVseqcsn4rpnLUX1t6E3BSgH vv36FrakLsfwGEAGG4reJ2WAdZR5DILLJzDQajEZ4zzKBhsYtLW3sP/0HlbMuZUJcVM87jt1oYyU MdND/faDz2tEh0YfgbJ3TzBzsTkG3ksnG8dfFXC0QVjinFDvDTHK+MsTCA/Tu5edBtePv8rzTRw+ cZ4jJ1xzJpw+19iv7Z+ra+FcXYt7sSitRsOUcXHMTBnj+puSxLSJCYQZdP3a3nD1hQYG6fe9hLrP q9CYxkMvKR58wYoFPSZjd5VSVjyKsj1EswVMJmOf5caUm3lN2WhutmAwmQY1d57MYxBcPvMYaLSY jDEeZeZBBgbHzx9BoSiu3OsRGDRbzWwqXM8/3/IUWo1nMmmN+SzJMeNDvVuGjPIKvHwDg8GPSlBO B9bqQo8y48TBdfkIESqazov5lHFxrLx6FgAtbe2UVtS6JleqqKH0ZG2/WhWcSlFxtoGKsw28W+ia Glyn1ZA6IZ4Zk8eg0Wo4WVVPxdkGIsINTJ+cyMwpY5g+OZEZk8eQnDA8W9SGzRRSeqMRk/87MJkG UI4ek2nw46WlKyG4/A1XNBk9uxJaOgODbUfewObowO6wYXfauWX+131WY+zp6LlDABRX7uXmeXe4 yw9UfkpdSw3Hzh0ifXyGu/xEzVG2Hvo7D67I9dhOk6WBprYGpiRO8yh3OB3otMP8l4BXi4HWENk5 OsDlYloMOi4cRHWY3bc1hijCky8P9TsW4qJFR4azaM4kFs2ZBLiuA2dqmiitqKX0ZC2lp2o5caau H7kKrlbn8jP1lJ+p9yi3ttvZe/gMew+fcZeZosKZPimRGZMTSZucSNqkRCYlx4b8B+mwCQyGC+lK CC5/wxW9AwOzpYm2jlbeLvYcDbJo6rVMik8NuO3j512BwXlzNeebqhgbOxGA/adcv3J3l233CAw+ Ov4e5TUlVFw4TuqYGe7y7Ufe5EzDSf7pxic8tr+95E2SY8Yzf/JVod6NAXnnGHjPQngxLQa+3QhL 0GjlFCJGHo1Gw+SxcUweG8eXFrvODR02B+Vn6jh6qpajFbUcPXWBMzWDz4cCaG5t5/POyZ26hBt0 TJ2YQNqkxM6/BKZOTMD4Ba4ZId9qL9KVEFz9CQyarU00ttX5PPdc4xnCdOFs2PU77E4bdocdoyGC x27J40Lzeepbu5Piis98xpdiJ9LQWsfJC65V3I5U76fJ0kBsRDxNbfUcPPMZAO8ffp3vXfcjwNVa UFj+AXanjQOVn7qDgCZLA9uPvEG4IYKZYy8jIiwq1LsywA7uKzAYXIuBcnTQfnaPR5l0I4jRJMyg cy8W1aXV0sHxStcES8dPXeDY6VrO1bVc1Ou02xyuloqKWneZRgMTkmJIm5jItM5AIW1iQtC6IiQw 8OKUFoOgcnpduLRaHdHhMWjQoDqz59s6WqhrqfF57jlzFRqNhprms+4ys7WRGvNZymqOeDz2QOWn fOmyLPaf7u4Tdyone8o/4Ka5t/NJ2XZ3kFJy9gBVDaeYGJ/C9iNvuidYenP/K1w2YQF6nZ43979C h6ODDkcHb+x/ha9f9Z1Q70r/vLtqDJGedzsHFxi0n9+HsnePUdSGxxE25rJQv1shQioqIozLZ07g 8pkT3GXm1naOn77A8dMX0Go1TJ2QQOqEeFos7ZRV1nH89AXKKusoP1Pf77UelIKqGjNVNWZ2fn7S XR5pNDB1YgLTJiQwdWI8UycmEKbXcbK6AY0Gd2vHQElg4MUpLQZB5a/FQKvVEhUeTUt796xmZ+or fJ57rvGMT2ABUF5TwrFzhz3KzjZWcqH5PJ+f9kyW23PiQzLTb6Gw/AOP8m1H3iBrwd0e5fWttXx0 /D2mjpnBvlOf9NjGByxKvYZpybNCvTt9KJ8WA6/AoMfFfSB8uhEmXI1GI7OCCuEtJiqchbMnsnD2 RI/ypPgopk5IcF+slVJU1ZgpO1PnChQq6yg/U0dji7Xfr9VmtXG4/DyHO9eO6GlScqwEBkNFkg+D yzf50HVxiTbGegQGp+tP+Dz3XNMZOvw0hR8/f4Tj54/4lL9/5HWqGjwn+2lsq+cvn75AS7vZo7y4 ci8d9naf6ZjfP1xAQpTvZBib9/6RH938S/Q6zzHMZxsrGR832efxx84dZuY4z1/YHfZ2Ss8WkzH5 So/yrhwJ7/J+8ZoSWReR6Ln/rQ0D36TTTkfN5x5lxknSjSDExdBoNEwaG8uksbFcv7A70bmusbUz edHVqlBeVUd1rRnVd96jh+paMx02+6CWqZbAwItvV0KoazSy+M5j4MryjzHGcq6pO1vXX4tBQ1sd bR2tPuUHz3zms12AvSd3+a3D/tN7fMoUipKzvlMFW2xtVDX6ziRY03yW94+8wc3zvuYuq2+p5en3 13HnovtYmHqNu/yzkx/x5z3Pc/eSB1iQerW7/C97XqC4ci93LX2ABSlLATBbGnnhg/+g0VLPXUu6 y/vLu8VAF+k5tbSj5+RE/WSrP+qRtKg1JhAWP3PA2xFC9C0xLorEuCiumtv9A8PSbqOiuoETVfWc OFPPiep6TlbV02YN3BXhVIrT5xqZPnnMgOsggYEXpyyiFFT+5jEAMEXEeZS3dfhP4Gm3+zaz+QsK vgjbS97g8imLGdc5+uGve/9Ah72dv+79I+NiJzExPoUa81n+9tmLAPzl0xeIjYwnLXk27x9+nQOV nwLw58LnCNOHMz15Ni98+J80dCZevrx7PU6nw+/00Eopv61ZzQ4bPTsPdJFjUQrKnDrSdHYcFt+k zr60ey2vHJaUMeBtCCEGLyLc4JP4CHC+rpkT1Q1UVNVzsjNwsNkdpE6IZ+qEBEyR4YN6PQkMvHg3 10hPwtDyl2MA+ExyFGo6rc5n+We91oDREOHuhnA4Hfxlzwv8040/Y+/Jne7uDLvDxoaP8vmnG3/G S5/8t7v7w+F0sGFXPl+au4qtB//msU/+9PEzjI+dTHXjaXe5QvGXPS/gcDpYnHadu7xr+1PHzOTG y25zl5fXlPI/9a3M1IVxU1gHOg2o8FjetoVzxK5jilbHLc4Wku3taPT9P2F01BZ73A6XwECIYWFs oomxiSaWzpty8RvrQX4Oe/HpSpC+hCHlb4IjwGfI4lALC3AhvHzKYp+y+MhEvn7ld33Kr09fydcW 3etRVll/gr999iKvf/5nj/KG1gv8x9u5Hhd6cHVNvP75y+4RGF0cTjtnGk76vKZC8de9f+Dj4++7 H/e/H/8XpWeL2XLwVV4ufA67w07p2WKe//A/6AAOOfRsshq54NSwYc8fOWJ37ePTTj3/azVy9Myn Pq/T0HqBT0/4dr04O1qwNZZ77ssxEhgIMZJJi4EXn64EaTIYUgG7EoYoMIgON3kkMXa57fJv8cb+ v3h0RSRGJ/OtxWs4VVdOQ485EG6YcxsLU6/ho+PvU9mZBBkTEc+KOV8lXB/O3IkLOVRV5H78nhMf +q2LxdbGUPl70Z9otpo513SGI9X73eVFFR9zrvEM58xnPFo4zistG6wRYD3msZ02tPxP4f/wpeZa vnTZKrRaLRUXjrNhVz4t7Wb2nfqEr1/1XRKiXP2SHRcO0XMRJn1MCjqvtS2EECOLtBh48RmVIC0G Q8rh3WKgHdrAIGPyVUxO8JzK2KALY0HK1Vzhlch39fQb0Ov0ZKbf4i6Lj0zkqmnL0Gg0ZF1xl7v8 1vlfJ7yz1eHrV32HmAFeHLtaRrylj5/vUzYmeizfuOoffKZffu/wa+5JmXqqajzl0+3RG4Xi3cOv 8cy2X/Lh0a08u/1Jd/fI8fOH+c8tj7O7bAcg+QVCjEYSGHjxnhJZJy0GQ0oFzDGIG9B2dFo9acnp PuUzx81lvtcwvzkTLifcYGTp9Ex3mUEXxuJpywFYPG25OzC58bLb0HVO8zs1aSbzJi1iSmIaC1K6 RxNEhZv41pI1fuuVkjjdJ2jQanT84w2PM2vcPI/yq6Yu5x+u+xFfnts9siE63ETO9T/hqmnL+M6y R3pdG2Ig4jRONF75M6fqynj985dxeK2v0G638upnG1i/4985f9ZzmGJ4kiuQaeto9Wn9EUKMDBIY eJF5DILLN/mwK8dgYMmH42InMnPsXI8yjUbD9OTZPuP/u1oKJsWnulsTFqZe7Z7WWK8zcH36V4iP TOTKqcs8nnvr/G9w+4J7fI6DmeMu4/r0r3iU6XUG7lqyhvuufcjj1/4t81eTOmYG917zA/ccBymJ 07lj0X0A3DR3FdenryRMF8b3rvsRidGuzOP08Rl8P/OfiQzznfY0IWqMT2Ck0Wi4OczONfoOjyBg VtwE7jNaWW1sJ0rT/8HQx88f4X8aWym0GXAoQKsnLHE2J2uP8R9v5/Lrd37qM7GUEOLSJ4GBF0k+ DC6fHIPOroSocJPPksi9mRif4nNhnJwwjYiwSBKjk5kYlwKA0RDB7B4LJ3W1Glw740sez106fQW3 XXGXu7WgyxjTWJ9VFrt8JeNOj0WdVs67kzGmsaSMmc7tC+4BXBf362atdNflH5b/iCkJ07jv2ofQ 67pf66uXf4uHb/q5TzdIypjp/ODGnxIX2T1RUVxkAt/PzGXN9Y+yeJprtIJWo+Oepf/IPL2Nq8Ps fM1oJRzF4mnLyZ53OwYNpOgc3BduYUaE7zoPep2BZTNv8knSdKBhl83A/1qNnI1MYWfZDp7d/ivM 1kbONZ3huQ/y+OOu33Gh2XfmNSHEpUkCAy+SfBhcgWY+1Gg0RIf7bzVIMo3zKZsYn8qUxGkeTe2z xnW3IHS1GsybtNBjdsIrpizhsglX+MxOGK4PH/BMgzqtnruXfp8wXRhTEtNYPuvL7vuWTl/BDXO+ yrcW53i0NsRGJvDQl/6VGK95GwD3fAjexsZM4KEbf8a42InEGOP4fmYuCdFJ6LR6vn7Vd/nq5d/k vmsfImPyIroSBafpnHw3op2vX/U9DFHdE5xEauHOGCN3LrqfsM59F2OM4x9X/AurFmTzk5t/xcyx vmsg1CktG2vO8vrnL/u0+hyu2sdTW/6Z1/ZtotVP4qfDaWf/6T202we/sqMQ4osjoxK8KKckHwZT oK4EAFNELGZro89z0sfPp7b5nEfZxLgUdFo9qWNmcPy8qzm7Z9dCxuQr2XLwVa6Y4plwGKYPD5gf MBjJMeO5Y9F9TE6Y6tPi8ZWM1X6fM5juqdjIBH5ww09paW9mjGmsx31dXRrK4TkLWpTOVR9thOfM Zw5LHUunZzJ9rGuipa/MX01sRDwACdFJrMl8jM9OfsRre56nvysrOJwOdh17h70nd7Ji9q3u1ocL zefZuPtZKutPEhVuIjP9K1w940Z3IqcQYviRwMCLb4tBqGs0sgSa4AgCj0yYPT6DXcfecd/WoGFC vGtCj7TkdI6fP0yYPpyUMWnuxyTHjGd68mymj53js72IsEiGkr+ZCYMhIiyq1+Weldc6CRqN6+ut DYsBrR46kwyVvQ2n3UKSaRzfWpLjd1tXjJtNYkQbH3aEcdChAzy/CEZDJMmmcT5rWlhtFt4u3syu Y+9yRcoS9pR/6B4i2trezJsHXmFH6dtcP2slV8+4AaMhAiHE8CKBgRffwEB6W4bSYAKDaUmz+Jdb f41CuZNDu35xduUZTE+e7ZMf8PWrvucz5G9E8x6y2KObRmdMxNHWnQfgtNShNU0KuClbYzkRGrg5 vIMMh5b3nSbO21wtEhPjUvj2tf9EYnQyn58q5K3iv3rMAwHQbG1i59F3/G67tb2Zt4r/yvaSN7l2 5pdYNvMmosJNod57QohOEhh48Z52X+KCoeW9bLK2x4XbX2AQFW7CoA8jITrJ7/amJKSh1xmY4adf PDHAc0Ysr/wNegRK2gjPwMBhuYC+j8CgywSdk5y0qzkckcr5pjN89YpvuXM7rkhZwrxJC/no+Pu8 f+R1LH4WuQrEYmvjvcMFfFC6haumLuO69JXuERlCiNCRwMBLb79oxcXz3b/dTdT+AoO4yIRet6fX 6Zk6ZgYzx81ltFNegYGmR/6GLmIMPTMQei6m1Fr2Ou0XDqI6mnF2mIlO/xa2Js9pkMPjp3PN5Ov8 vq5ruOdKFk9bzo7St9l19B2P5bENujBuvfybtHS2IngvhGVzdPBx2TY+Kd/O3IkLWT7ry0xLmhXq 3SnEqCWBgRfvOVtkccWh1Wvyod/AILHPbS5IuTpgRv+o4t2VoO0ZGHjuR6e1OzCwVH2Evak7V6D9 3F5sjZ65A4a4NPoSERbFVzJWs2zml9l25A12l20nOWY82Vc/yNiYCQAsn3UzHx17j13H3vVZQVMp xcEzn3HwzGdMik9l2cybuHzKYo9RJUKI4JPAwItMcBRcvXUlxEQMvMUAYFHqF5P8N9wpp3fyYY/A wOg9MqE7J8DRVuNxn/XcXnB2ty9odEZ00RP6XQ+TMYZVC+4mc/ZXiAqL9riwR4ZFcdPcVVyXvpLd ZdvZdewdGtvqfbZxpqGCP+95ntc/f5mr0q5jadqK0dc1JESISGDgxaFkSuRg6q0rIXqQLQZaadZx 8U6Q6RF0ab1aDLq6Epy2NpTN85d7z6AAQB87Fc0gutS6hkD6E64P5/r0lSyb+SX2n97DjtK3OdtY 6fO41o4WdpS8xY6St5g1bh5Lp2cyZ8IVoyupVIgvmAQGXmQeg+DqrSshZhA5BqJboOGK4NuV0NVi 4N1a4I8hblqfjxksnVbPwtRrWJh6DWXnj7Dz6Dscqd7vsyw1wNFzBzl67iDR4TEsmnoNV01dzthe upAsHW1DPjRViNFAAgMvTulKCCrfmQ+7A4OIsCh0Wr3Hoj4SGAxArzkGnl0Jzs4Wg/4FBn3nFwyF 6WPnMH3sHC40n+eTsu18euJDv0tXt7Sb+aB0Cx+UbmFywjSunHotV0xZQmS4a00Jm6ODdw+9xgel bzM2diILpizl8pQl7qWkhRC9k8DAi/daCTppMRhSPmsleAVey2Z+ibb2VpzKiVM5/U6HLPzzbTHo 0ZUQZgJtGDhdowWUw4rT1jqsAoMuY0xjue2Kb7Fy3h3sO72b3WU7qPSaSKlLZf0JKutPUPD5JtLH z2fWuLnsOvaue6bMs42VvNVYyVvFfyUlcTqXT7mK+ZOvIlYCTiECksDAiyQfBpdPV4JXX/FXL/9W qKt46XIGzjEA0EUk4GjtnlraYanrMzDQ6IzoovqfeDiUDPowFk+7jsXTrqOq4RSF5TvYd6oQq59W BIfTweGqfRyu2hdwe6fqyjhVV0bB5y+TOmY68yZdScakRR5zZNS11PLavpc403CKuROv4LKJC5ie PFtGRohRRQIDL7KIUnAFWkRJXLzecgzA1Z3QMzBwWi70GRjo46YNi+B4YnwKdyy6j9suv4viM5/x 6YkPKaspGfT2Ki6UUXGhjDf2/5mJcSnMnbQAp3Kyo+Rt7J3Jl5+UbeeTsu2E6cOZNW4ecyZczuwJ 8wPO0CnESCGBgRdZdjm4ZMrpIOolxwD8LKZk7bvF4IvuRuiLQR/GwtSrWZh6NQ2tFyiq+ITPKj7y WWQL4OrpKxgbO4nPT+2m4sLxgNusajxFVeOpgPd32Nvd8ysATIpPJX18Bunj55OSmCajYsSII4GB F+lKCK7e5jEQF8l7SmSNV1eC0XNkgr2l2icw0OjCUY7u5ZENscEbkXCx4qPGcONlt3HjZbdRWX+S fac+4fNThUQYIll91XfcsydeO+NGGlrrOFD5KQcqP+V0XflFve6ZhgrONFTw/pHXMRoimTF2Dunj 5zFz7NyAU3cLcSmRwMCLtBgEl3QlBE9vUyIDGGJTPW63n/sM5eienlijMxJ35U9Q9lbQhaPRhQ+7 FoNAJidMZXLCVL46/1s4lRO9zvPUFh+VyPXpK7k+fSUNrXUcqiqiuHIvJ2uP+QyNNBljyZx9C7Xm cxyu2ud3KfAuVlubR2tCYnQyM8fOZcbYOaSNnU20LA4lLkESGHjxigtk2eUhJl0JQdRHV4IhwXP9 AUfrWY/busgkwpPnh/pdXBStVouW3o+p+KhEls28iWUzb6LFauZw9eccOlNEeU0pV027jpvnfc29 HPQdi75NVcMpjlR/zpHqAwFHR3Spa6lhd8t2dpdvB2B83GSmJ88mLXk2aUmz3EMqhRjOJDDw4p01 L10JQ6u3eQzExekz+dCYgDZiDE7LBb/P10WOvpUNo40x7pEPTuX0CVQ1Gg2TElKZlJDKTXNvp8Vq 5ui5g5ScLebYuUO0tjf3uv2zjZWcbaxk17F3ARgfO5lpybNIS0pnatJMYiLiQr0LhPAhgYEX7xFf Mo/B0PLJMZAWg6HTR4sBQFhCOtaqj/w+XRc5NtTvIKT6cyxGG2PcMzUqpTjTUMGxc4c4eu4QFReO e0zO5c/ZpkrONlXy8fH3AVfXw9QxM0kdM4OpSTMYGzMx4I+RY+cO0WI1MyUxjTGm0f1ZieCSwMCL JB8Gl09XgmR0D5m+cgwADPGzegkMJHFuIDQajTu34YY5X6XD3s6J2qMcP3+EsvNHqGo45Xdq557q Wmqoa6nhswrXZ2I0RJKSmEbKmOmkJKYxJTGNGnM1bxdvprym1P28qLBoJiemMSVxGlMSpjI5YRrR xphQ7xIxQkhg4EWSD4NLkg+DqF8tBrMCPn20txhcrDB9eOcwxgwA2jpaOVFTSllNKeU1JZxtrOwz ULDa2txrQvSmtaOF0rMHKD17wF0WH5nIpM5AZVJ8KhMTUiX5UQyKBAZeZIKj4OptESVxcfrKMQDQ x6b4DEnsMhpzDIIpMiyKuZMWMnfSQgAsHa2cqD3GydqjnKg9xpmGkzi8g7mL0NBWR0NbnXuEBLhW J50Yn9L9FzeFeFkzQvRBAgMv0pUQPD6JnWhk/w4l74uMn9YYjUaHIW46HXWHfe6TwCC4IsKiuGzi FVw28QoAbPYOTtefoOLCcU7WHqOirgxLR2uv2xgXO4mG1gu02639eUka2+pobKvzmCo6IiyKCXFT mBA3mQlxUxgfN5lxsRMx6MJCvYvEMCGBgRfpSgge7wWUNNKNMLS8JzjS+v96GxJm+QQGGkMUWoMs UfxFMujDSEtOJy05HXD9KKltPkfFhTL3ug7nms6glCJ1zAy+krGatOR0nMpJjbma03Unqawv53Td Sc42ne5364Olo5XymhLKe0wprdFoGBM9lvGxkxkfN4lxsa6/MdFjJQ9oFJLAwIt0JQSP8llASU44 Q0l5XRg0AWaVDEtIx/t3qbQWhJ5GoyE5ZjzJMeO5atoyANrt7dS31DI+bpL7cVqN1n3h7nqc3WHn bNNpztRXUFlfwZmGk5xrOtPvYKErKKltPkfxmb3ucp1WT3LMeMbFTmJszATGxU5kbMxEEqOT0cms pSOWBAZevFsMpKl76Di8ftHqJL9gaPUxJXIXQ/xMnzIJDIancH24R1AQiF6nZ3LCNCYnTGNpZ5nd Yeec+QxVDaeoajhFdcNpqhtP97sbAsDhtLvnYuhJq9GRZBpLcswExsZMIDlmPEkmV1DTNTmUP81W M4fOFGGKiCUpeiyJ0WN9ZqkUoSefiBelpCshWHwnj5IWg6GknN7Jh/4DA21YNLroiThaqtxlMiJh 5NHr9EyKT2VSfKq7TClFXUsN1Y2nOdtY6fq36Qx1LTUD2rZTOThvrua8uRrv8RMmYyxJpvEkmca5 /mLGERUWzd6Tuyiq+MS9eiW4fnjFRyYyxjSOMdFjSTK5goUx0ckkRifLctchIoGBF+lKCB7vwEAn gcHQ6meOAYBx/GI6ag+i0RvR6I2Ej5kb6tqLL4BGo2GMaSxjTGPJmHylu7zdZuWcuYqzjZWcazrT +VdFs7VpwK/RbG2i2drEidrSPh+rlKK+9QL1rRc4xiGf++MiE0iMTiYxyhUouP6SSIhKknkbgkgC Ay+SfBg83rMeSovB0OrPBEddTLPvgtmhrrEYLsINRtfESomei2a1tjdzvqmac+YznG+q5ry5ivNN 1b0uLDWUGtvqaWyrpxzfICNMH05CVBKJUUnER40hMTqJ+KgkEqLGEB81hsiwqAG9lsPpkLyJThIY eFE+iyhJYDBUfOYwkC/h0OrHBEdCDERUuIlpybOYluw5MZalo42a5rPUmKupMXf+23yOupbzfSY8 JkSNITE6mdrm8zS21Q26bh32dnfrhj/heiPxUWNcgUJkIvFRY4iLTCQ+KpH4yERMEXEo5eTgmSJ2 l22nrKaEqHATcZEJxEcmEuf+S+j8N56YiPhRETxIYODF6TOPQahrNHJIV0Jw+U5wNPJPYCI0IsIi /bYwOJ1O6ltrqW0+S03zOS50jnSobT5PbEQ816XfzLyJi9wjkmyODi4013Ch5TwXms9xoeU8dS01 XGg+T2NbfZ8zRfam3W7tNXDQanTodXo67N2TfbW2N9Pa3kxVwym/z9GgwWSMJTYygbjIBGIj4t3/ xkbGExvh+r9Bf2nPCSGBgRfpSgge6UoIMu8VwEbBLxsxvGi1WncOQ396qgy6MMbHTfI78sLusFPf WutaT6K1lrqW89S11FLfUktda43HBX0wnMpBh31gM08qFGZrI2ZrY69LcEcYIolxBwtxxES4Whti I+KIMcZh6vx3uI7IGJ61CiFJPgwenxYDuXANLa8WAzTy9RaXLr1O757XwZ+W9mbqW2qpb63t/PcC 9a21NLReoL7tAnaHbYCvOHQstjYstjbOm6t6fVxkWDQxEbGYjHHERMRhMsZ2/2uMxdT5Fxke7ff5 jW31fH5qN2ZrI6bwWEwRrsdHG2Mwhcdgiogb1Ho0cubwIvMYBI/vOgnSYjCU+jvBkRAjQXS4iehw E1MSp/m9v9lqpqH1Ag1tF1z/trqmh25oq6OxtY7WjhbAtaLlotRrWJJ2PRFhUa7HtNZ1Jj7Wdf41 0NhWR0u7eUjfQ1tHC20dLZxr6j2A0Gl1RIfHYDLGYTLGYDLGUtda47Hipj//cuuvSYge+KqpEhh4 kXkMgkfJPAbB1c8JjoQYDVwX0JiAgYPN3kFDWx3xkYkeOQFxkQmkjpnh9zl2hx2zpcEVNFjqaWpr oMn9r+vPbGn0WUX2YjmcDvf2B2KwQzolMPAiXQnBIzMfBtdAhisKMdoZ9GEBuykC0ev0JEQn9for XClFS3szTZZ6zJZGzJYGmjr/dd12/TW3N/n8EB1KYfpwwvThg3quBAZeJPkweGQRpSDzmvmwtwmO hBDBodFo3K0VxAd+nFM5abGaabY2uYIFayPNlib3BFE9b/c2jfXkhGnMHp9Bu93a+VzXNgcbFIAE Bj5k2eXgkUWUgktaDIS4dGg1WmIiXEmHE+NTen2szd6B2dpEi7XJffG3OdqZM+EKxpiGfjpzCQy8 SFdC8EhXQpDJBEdCjEgGfRiJ0UkkDiKRcDDkJ5sX6UoIHkk+DC5pMRBCDAU5M3uRroTgcTi95zGQ w29ISYuBEGIIyJnZi8O7xUDigiHjPYRHK79oh5TvlMjSUyiEGDgJDLz4zmMgu2ioSFdCkPm0GMj+ FUIMnJw5vDh9VlcMdY1GDocsohRcPhMcSYuBEGLg5MzsRZIPg0cWUQoumRJZCDEU5MzsRZIPg8d7 38oiSkNMFlESQgwBCQy8yDwGwePwST6Uw28oKWmREUIMATlzeJGuhOCR5MMg884xkCmRhRCDIGdm L9KVEDwyj0FwyQRHQoihIGdmL77zGEhgMFRkHoMgkwmOhBBDQAIDL77zGEhgMFSkKyG4lE/yoQQG QoiBkzOzF+lKCB6ZxyDIvJe1lhYDIcQgyJnZi3dXgk5aDIaMzGMQZDJcUQgxBOTM7EVaDILH6d1i IL9oh4x34iFo5dgVQgyKBAZeZErk4PEODGQegyEk6yQIIYaInD28OL36aWURpaEjyYfB4ztUUboR hBCDI2dmL149CUhr7NBxeP2qla6EISRDFYUQQ0QCAy8+LQYSGQwZ6UoIHpncSAgxVOTM7MUnx0CS DIaMdCUEkVPmMBBCDA05M3uRUQnB472Ikk4uXkPHK+iSrgQhxGBJYOBFFlEKHu9uGmkxGDrKq8VA IwsoCSEGSc7MXmTZ5eDxncdADr8h4z2PgbTGCCEGaWSdmZtLeXrNCjQaDZpVj7On1j7gTUhXQvDI IkrB49NiIPtWCDFII6i90cqrD87m4eZ8Sqpe4NAzX2NJchw16lGSBrAV7ymRt+z8lcwVM0ROmGs8 bref/5xmZ32oqzUiONsbPQskx0AIMUgjJzCwHuHZjVBQ9RDpEyD9iefJeDKH0oZHSYrv/2a8Www+ rDkucxkEiaO2iNaGT0NdjRFJJjgSQgzWCDt7ZDLW1Plf4zTupJjdJxtYFt//yODeWxagFGw5+HdQ GgkKgkh2bfB89MluVl0ne1iI0c77x25/jLDAoM7jrcVmQJNtYFu4e+UVAHxu/m2o38yIJz00wbNs 2XUo9UGoqyGEuASNnMCgMwBo7crBsp7ktWL46bQB9CP08OXJGYgT+3EAABPISURBVKF+RyNeWux4 TBExoa7GiKSLGEhmjRBCdNOowbQzDEsNPDc/gWfv3knRo8soe/URZq+GCvVbUkJdNSGEEOISMYIC A7Ce2so3U1dSAEAGG4re574F8stJCCGE6K8RFRi42GlutmAwmTCGuipCCCHEJWYEBgZCCCGEGCxJ DBdCCCGEmwQGPQ3BlMqiB3s1Lz++yrU/V6zh1YO17rvKtz7NCo0GjUbD4y/vQfb0IDQc5Ber5qPR aJi/6hfs6zxem8u3sma+a9+uevxlqmXnDpi9dg+Pd+7bFfc8xb6Grp3YzNan17iOac0qXt5THeqq XlqsB7lHcw/7rF0Fdva9/AvmazRoNCv4xav73OcCOUeEkBKdLGpzNoqsfFVSVaY252YoyFM1oa7W JawoP1ORsU4VV1Wpwk1rFWSpYotStorNClD520pURdFmlQFq3U7Z0wNTo/JAZawrUDX1VWpTDoqs zcpmK1M5oLLyt6mKiiKVCyojb2eoK3uJqVC5oDLztqiKqhK1PgdF5nplVkqVbc5RkK22lVSoos25 CjLUzvpQ1/cSYKtRWzbkq+wMFGSpQrOruGbnOgUZalNhmSor3KQAlbezRs4RISaBQRdLkcoEVVDV dbtQZciX/iKY1YbsTLW+2NJ5c6drf5qVKl6fqcjZ4n5kUV6GYl1hqCt8SbGUbVCQqypsSpnr65XZ XKPKyqpUQ/F6BWtV12FsLspTkKfkMO4/W1WBAlRh56HbdS4oNJvV+kxUjvskYVZ5oNYVyt7tk61C bV6/QW3asM4jMCjKy1DkFbkftiUHlZFXKOeIEJOuBA/+p1QWg2Hivpe2s2YebH/6ETQxyynO+RlL TWBrh4y0ZPcjJy+9E17dQ3Ooq3wJsbW1A0+SatAQk5BATMyN7GqKQEc7MJWuw9g0eRGwlWOyc/tN bxpLBlBe5Wrvbjiyn2ISAWivg8smufcuS9dl8OqHJ0Nd5eFPn8Kda+7jruxVZGJ2Fy/44R5sjy4A wH5qKyufhwdvmyPniBCTwMCD75TK7QOcUll405Ny7dfYnJ8Dz/+C18qtGIBiunesMSociq3ShzgI G4rqUUpRsnkJ9y98hnMA9NiXxigyPY5r0SfTQp7Jy+Tu6RGsWLWChIUPADEYOu+29nhoVKznsSz6 YPfaV3ojeqzse/lxDKkrydlQxJp0k5wjQkwCgy4BplS+ZpBTKo961nKefvxpSq160hYs486H/pv1 GcWUXLC4dnVjh/uhFR9thdxFyJ4eAFsTkMcdC1x7Lf3WNWTwKicaAdrdJ1BrxR52cCfTTIN7mdHq yocKqCor5ue5+dTUFJKJ2X2ZamzsCg2a2fuHYnIXTQt1dS9htTy3KoKFd1vYWWHmuftcrQdyjgix UPdlDB/1an2GK1HLppQq2bxWwVpVEepqXbLqVT6ozPWFyqKUqi/bojJA5ReblbkoX0Gm2lllU8pc otaCyimQPT0g5kKVCSpvW4VSyqJ25mcqMvJVvdmVK5O3s0opZVab16LIKVC2UNf3klKmst37UKnC /ExFhiv5sKhzP7sOXVeCXEGF7N1+sxSpTDLdOQYlG7IUrFNlZpuyWczKXG9WFpuSc0SISWDQg6Vi i8oCBSjIUBuKJAv2YphLCnrsT1TO+p3Klc9lUdvystzlGTmbZPTHIFRsyXPvQ8hSBWWus23Ftp7l a1WR5MYNWJnHvs12XaCUUspSpvKyehzTG4ou7oVGG0uRyiRLFZmVUsqsNvTYl11/+UX1Ss4RoSUz H/qQKZWHmtVqBb0Ro9dannZrMxYMmIyypwfNbqXZYiPCZPJcKtVupdkCJpPs20ELtG8Ba3MzGEzI oRtcco4IDQkMhBBCCOEmyYdCCCGEcJPAQAghhBBuEhgIIYQQwk0CAyGEEEK4SWAghBBCCDcJDIQQ QgjhJoGBEEIIIdwkMBBCCCGEmwQGQgghhHCTwEAIIYQQbhIYCCGEEMJNAgMhhBBCuElgIIQQQgg3 CQyEEEII4SaBgRBCCCHcJDAQQgghhJs+1BUQlzbrqe08/afPILxnaTvtsdfwyO2RPPu7/XztiQeY bgxuPUpff44/HYB7164h3dRrjXn1kcW8d82f+Eb1O3Tc8hA3p/lWzl67h9/8dj93PLGGnndXb32c 7HeW8O5vbxsdXx7rKV58+k/UeH7AtDfC9T/Iwbb5edc+DMoHbGfPy8/y7hmrx6vHJk/j6ltuZV5S gNds2MM9CU9yb/FaCv98lLu8PsMu1bte5A9H0vjJmmX0vHu0fca9fYd/suZqPsl/OoifsUvzqT1s eO6PvPZWIXUksuLOe/nOA98M/Bn383sc+AX3cc+1z/HonueYF+Rz06VIo5RSoa6EuHQ173uamIUP k5GVzZKxka7CtjbaJmfx1J1VTFj4B3Y3F7MkOoiVsJeyxjCb54G1BRX89raUwPU9+BwxGVsosrzE RxExNBXW87PF8QHe1x8oNB9gscnzte4xzObaYjNr5vUagYwMzftYEbOQHZnZ5MyIdBe3nYfVv/5/ nJyeRNPuep5YEn8RLxLwxXluRQwP7MgkJ2dG5+seZ2PBDiCLnTWvsSzJ91m7fqFh+bHNNDxSTby/ z7DTvqfns/Dh71KvHsKj9qPsM+7tO/zsr27mjxpTED9jqN3zHMlLHgCyyNvwdSa1l5P3wBMUk0FB xR5uS/G9cvf3exyYna2PGFiZsBP1s2Vf/E4f7pQQF8FcvF5Bpiq2+d7XXPx7Bderoubg1qFm2zoF mSonG0VGvqoP+EibKshBkbtNKWVW+Rmo/GKL6x6b5xvoel9FZt+tFK7LUGRvVjY1CliKVCao9Z37 yVOz+t08VF5R/YA32z9mtT4Tlbm+2LO0eJMCVL6/17UUqyxQ+UXmzs8wq/PYtCmvj1gVr89UZKxX fj7iUfUZ9/YdDv5nXKFyQZGxTpX1fP36QpUFihx/n0H/v8e9sZRsUJDl9zs+2kmOwShR29BKWeWF fv9Vnm8cwNZj+vWo5vKtPLJCg0bj+nvkue00A6e2/gLNil9Qbnc9rvz1x9HMf4R9za7bDfueY/78 xym1+t0qbz39BOT+lH//t01Q/DBbyq0BKlDM756H9d+6EoDwRNi75XnWzNdgMBjQzL+Hl/dUezzF 1vnvqe1Ps0Kj4RevlzN/9SOwcSMlAV4mFKw2C2caKgb053A6+r399n5V4hQvP36P+/NdseYpDjbY oXkfazTzefFg5wdau4t7NPN5eldt5/MO8kjP+/tgmjQu4H2n3v0jBeRwU0bXL/0C/vjkI8zXGDAY NKxY8zSlfl+mmdd/cQ8azT1sr7YPy8/YYW3A1nSi/3/m0wPY+sV9hwf7GVtLd/AksH7DWtJ69tvE L+bfN68jOxEsPpUY2Pe4uXx7jzqv4Lnt5QAY05eRQwFv7a8N5cc6LI2GLjQBvPLuAV7fWdLvx8+Y Mob/fiyrn48uYMtrW6mNhg6ADkhddiOTej6kdjvXTl9JceY6thXdQkfxq6y8/wa2N+2k6I7ZsGM1 u8p+Qlo67Nn8JBTD+0eeYMHieD5/9QGKp24i1V9fYPXH3F8AeblXEp9Wz1og7+W93OWnebC54iN2 kMl/ppqAZsJjYONjD7N2wzbKlkTx1v9bwt1LNhJWYeHLAMQQGeEKVKZnPUn2+kJyb0tDb20ik/v5 5Hgz84ZJU3PFheO88OF/Dug5/5r1NDERcX0+LgY4vONNttdG09H5AUdNXMqyeRE99y4vfjOV+wsy WL+lkCujKvnt8tVkPN9Ime0HTKKY3245wn3zFlNdtIWNFLNxSxEPLbuZ5iM7+B2JFE7yvy937NvB 9u21dHRAGOd4f/3dQC63ZPg2HddVHICs75Gq7w7qfvfEATbtLGZB2FH+eclqZhdaqTnwqOvOxHAM NPDimgTufz6DzSUfsGKCHhIyht1nbDn1Pi1HX+n347URY0j+0vp+Ptr/dzi95y7u5Tt84NG0QX3G xwu3AVlcOcd3H6ff+TNeutO3pgP5Ht8ZuZes6TdQt3YDRc8u5My7T5F1w3QoNrNmXgq3r4WVWw7y s2UrQvvhDjMSGIgh8djqlR631xebubvH7YN/f5picih++2euZJ8FCyimkoz7f03x918gD/jTh8e5 L93Alo2u52z98BiPLp7MG0/C2i3X4y8uOPjGeiCbG2dGAGP56roMfvfE8xz8yTKfpCJroxWYiMHg ut1uBnK38Nv7XCeFh16q4NzGVJ4tOMKXMwFO8ucn1/DkE8+TW1DGr25Lcz3ROI5MYN/RGhgmF41g MgHPP7ya53uUZa8vYtm8md37tvwN7i+AddveZ82KJGAxGyoK2JiaxVvFP2HV+iyeeOBdGh5dTOnO t1xPenIn1b+6mXPv/wGyHmO+ny7i8Bjg+Ye54Xnve+qorrKS5tH/3Mze13aQueo/MdIdGKwvKuCu BSZgHn8p3kBExv28V/4Q84iBHX/i4Xt+y/MbM9hWc4AVXTkLo+wzBv/f4fR4jft2b9/hfd9/bVCf sa29GZhK7ACuRAP5Hs8K/zU7yKbwJ1lMjYAZ9/wL6/6wkWdf28+aecuYmJ4Fzx6i4VcrCE4GxaVJ AoNRIudri/lO1qJ+P16rHUgvUyZFlu0s8LoQtxzs/r/BZALSPH71z1hyA/AnbCSxcn0Wj71SxKkb w9jIOgp3hrNk+R5OfeM8vyOTwmsm+HndU/z5gQIAFiZs7FFezJ8/yWPeCs/nnCv/BDJWdNehDtat XNzjETFEeDyjmCefKAag8MAp7LeldX5hTEzOhFdPXADShvJjGrSZY+fy/+54bkDPCdf3Lx27Csgv NvOQzwWypfu/tg4gk1uu7M4G1Cd0n2pn35wNrObzU3fx/pPFbCjaye6Fy/ns1BoqHisOGPi1myEz v5jtD81zl9kb9vGPCQtZ/rvbUb+9uY/aZ3F1j1+jxtlLyAbONFmYhwko4PnOQ+f4iQZWJHXVefh9 xlEzVhGZdusAnqEZwGP9f4d7fsa9f4chYxCfcWziWOAAFyyQ5nV4ndr6FA+/M4kNv73r/7dzP6GN lGEYwJ/WZm2FCPvXQ4XF2yo6C5XiXhbNroKLuHOVJccligfNniQeeohgSS+bFsEoSA42BwnCzimL kFaCh/EQwXSxhx3YAZnVTW0KDexEJvB6mMl0ko5p0rK77eb5wQdtJpOZ6TvffG8z7zddg/Yw/TgS fQHAt7gwvdz94S/9gS1cBLANnHyWA2EP1hiMiGORZ/Dc5LGB2+SxIbuK03/xw38sAH+hHnjNuvO7 //PL78aB1ZvILmSgpN7CG7NvI4bvkM1mAfU6zof807b1q4Z5AIuVdVimCcMwYZk1pGPA/NJP6L2V /KJyCait4G5nwUlg7uc7gXe4RQ6brZ2D0UwHVimF1bnLmF/p3Its4s9V4MK5U481hv2Mj49jMjI1 VBsbG2LgcNoDvGkVv9wN/NVbDhQALQATZ99ERgGWsguYRwIXZ2bxjgpks1l8ihg+CE38wk0cfwUz CoB7f2PvqgQNa+ZOoUDbuoNlAGciEwCaANKwxEYxDnx04RO/ruUwxnhsPILxiakh2pDz8A7Yh/cT 47Oz7wFYxVc/rvUs2YC28Bm0e+hJ1ofrx07zAdwYC8Rrpq6hNPe+l2w8D2z+i0HO7pHypKsf6Wjr V70fnJVgG0UBIEoyL+umJbVyThRAkNS8quO6ZBQIAEmX69KpSAcgqZIVsmVbionwWQhGMSEARDO7 K5Rto+Duqy0isi151dueVpV6wxItowoAydUaO8dlu9sqxCGAKvq2+JX6i6NQzuwfa1hVerBivRO/ uGhVQ8z1iqRi7u+dCQ21fFwQqDQ3Cm6cEC+IHbpx7xxI5kWvVkXXddF1XYqL7npJzdy1RnUxJoi5 Mw3cGEKAhJTXLbFMXVIKBEiK4fTMSghUwdtdx/30x7hfHw7GeO8+vJ8YO6IlFTeeuZIYVl3qZk1y Cfe1dKW+a41h+rFjaQJA1LQmZqMhpu7OaFELhnTPbqAgJgZ0IJ0pYWEXFXc60FV/uqJZyUkM8C7W kFgiL8Gxu5ZPCBCTsnctqObi7mAcNibZVVEBSWrG7mWNisQAUXumuYljSAKQjL6TeKipVNc+pQq1 wL4HjquhSxwQJDTZqOUFiEnlUc3gOky8v/P/JQZfB6ey1ateMuA1JS6asXNiOKYmCiDJohszxxto 0qGJn4jItpeQ9TZFkrlK6EDjT1G0OzGExBNqYN24lLx9quVUP4kQEbHKaQEgmUpdtkcoxv36cG+M 9+rDw8dYRKQhWibRE+OY5MpG+NuH6MciImZ50U1gvKamilIPfE6iaAh14wOO6DFro9m0gcgUopOP /87e2jeXoJSuw7l1LXBf0d2nyFQUg+zS7RtjuILSAPe3R1Or2YSDCKLRJ/FIufv4YmwajbKFm4Ea k3arCdsBpqLRge4nM8b9PKI+3Gpiq9lCGxOInj6OfmfP8P149z5v/baAE6+vYd35HudYZNCFiQGN Fu9JftcNG9eGeYSqp33/NiLTV6CZDq6e5dXkMNpY+RxnLgOWfInBKxd2MMZHwAH7MbCFpfMncOvj GlY+fG0f6z/dWHxIoyU6g+VKAbj/YF+rN7eBQmmdA8YhdvrSHCqFV7G5tb+SMsb4CDhgP0argVM3 NPzApCAUvzEgIiIiH78xICIiIh8TAyIiIvIxMSAiIiIfEwMiIiLyMTEgIiIiHxMDIiIi8jExICIi Ih8TAyIiIvIxMSAiIiIfEwMiIiLyMTEgIiIiHxMDIiIi8v0HIyR0OQ3BvHkAAAAldEVYdGRhdGU6 Y3JlYXRlADIwMTEtMDItMjJUMTY6NDA6NDctMDg6MDCVSj71AAAAJXRFWHRkYXRlOm1vZGlmeQAy MDExLTAyLTIyVDE2OjQwOjQ3LTA4OjAw5BeGSQAAABd0RVh0dGlmZjphbHBoYQB1bmFzc29jaWF0 ZWSMKAOzAAAAFHRFWHR0aWZmOnBob3RvbWV0cmljAFJHQrMgSd8AAAAWdEVYdHRpZmY6cm93cy1w ZXItc3RyaXAANjLgOdUuAAAAAElFTkSuQmCC --Boundary_(ID_UQnfOsK19lMNJYbtgzxRtQ)-- --Boundary_(ID_+ckLzYkDTicPnCgDPFpFXw)-- From gorry@erg.abdn.ac.uk Wed Oct 19 00:26:33 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 48C5221F8AA8 for ; Wed, 19 Oct 2011 00:26:33 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.524 X-Spam-Level: X-Spam-Status: No, score=-102.524 tagged_above=-999 required=5 tests=[AWL=0.075, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JsjXQ0f0yyxK for ; Wed, 19 Oct 2011 00:26:32 -0700 (PDT) Received: from erg.abdn.ac.uk (dee.erg.abdn.ac.uk [IPv6:2001:630:241:204:203:baff:fe9a:8c9b]) by ietfa.amsl.com (Postfix) with ESMTP id 83C3821F8A67 for ; Wed, 19 Oct 2011 00:26:31 -0700 (PDT) Received: from ra-gorry.erg.abdn.ac.uk (ra-gorry.erg.abdn.ac.uk [139.133.204.42]) (authenticated bits=0) by erg.abdn.ac.uk (8.13.4/8.13.4) with ESMTP id p9J7QOnT029358 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT); Wed, 19 Oct 2011 08:26:25 +0100 (BST) Message-ID: <4E9E7BA0.8060406@erg.abdn.ac.uk> Date: Wed, 19 Oct 2011 08:26:24 +0100 From: Gorry Fairhurst Organization: The University of Aberdeen is a charity registered in Scotland, No SC013683. User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:7.0) Gecko/20110922 Thunderbird/7.0 MIME-Version: 1.0 To: Joe Touch References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> <4E932F77.5060006@isi.edu> <4E993917.5030508@erg.abdn.ac.uk> <4E9DD306.4020402@isi.edu> In-Reply-To: <4E9DD306.4020402@isi.edu> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ERG-MailScanner: Found to be clean X-ERG-MailScanner-From: gorry@erg.abdn.ac.uk Cc: Janardhan Iyengar , ledbat@ietf.org Subject: Re: [ledbat] New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: gorry@erg.abdn.ac.uk List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 19 Oct 2011 07:26:33 -0000 On 18/10/2011 20:27, Joe Touch wrote: > > > On 10/18/2011 8:29 AM, David Ros wrote: > ... >> Hi all, >> >> Sorry if I'm getting this totally wrong, but: is it really >> *necessary* >> that LEDBAT's INIT_CWND be *smaller* than TCP's? As far as I understand, >> the point of LEDBAT is to be less-than-best-effort over (relatively) >> long timescales, or at least over time scales longer than one (initial) >> RTT. And just starting up as TCP won't make it *more* aggressive than >> TCP. Is this correct?? > > No, but it could make it a lot like TCP if the offered load is in short > bursts. If that's not the intent, then the INIT_CWND needs to be smaller > than TCP's. > > Joe > > I think INIT_CWND should not be significantly bigger than *deployed* TCP INIT_CWND. (It may of course be desirable to be smaller or the same, and that would benefit in the way Joe suggested). When I suggested "4" could be OK, this was only 1 larger than current usage for a 1500B MTU, and equivalent for some smaller MTU. That to me is not "significantly bigger". It would seem OK, because if this induced congestion LEDBAT would react within INIT_CWND segments in a conservative way. If we care about LEDBAT being conservative compared to other TCP sessions, then I really think we should not track future new proposals to raise INIT_CWND. I suggest this would have side effects: - It could make LEDBAT more aggressive than *deployed* TCP implementations, that I think would be bad. - It may require LEDBAT to implement additional algorithms to ensure it is conservative when the larger INIT_CWND induces congestion. Addressing this would likely add complexity to LEDBAT and make it dependent on these TCP updates (if any). I think a larger INIT_CWND (e.g. by tracking any evolution of TCP's INIT_CWND) is unwarranted if the goal is for LEDBAT to target bulk less-than-best-effort use. I cannot see the case yet for a LEDBAT INIT-CWND beyond 4. Raising this could indeed save an RTT at the start of a LEDBAT flow, but this would be at the expense of making it (slightly) more aggressive than presently deployed TCP. This seems undesirable. I also suggested that MIN_CWND should similarly be fixed to a small number. Gorry From arjuna.sathiaseelan@gmail.com Wed Oct 19 01:35:01 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3E73B21F8B86 for ; Wed, 19 Oct 2011 01:35:01 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.977 X-Spam-Level: X-Spam-Status: No, score=-2.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id EihJMsyftJXl for ; Wed, 19 Oct 2011 01:35:00 -0700 (PDT) Received: from mail-qy0-f179.google.com (mail-qy0-f179.google.com [209.85.216.179]) by ietfa.amsl.com (Postfix) with ESMTP id 9238821F8B71 for ; Wed, 19 Oct 2011 01:35:00 -0700 (PDT) Received: by qyk31 with SMTP id 31so1224965qyk.10 for ; Wed, 19 Oct 2011 01:34:59 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:content-type :content-transfer-encoding; bh=qDNnV6GsQQdebu5Xtdp4bqKG1j0QR1MGNSnYTtI4VTM=; b=IZV+1SBVrDY9B7XDFrGUU5aLP94ByZaPPsqhCs/rXmV9VxHnYdLf7nfnSC49REwWih NY721nr1SsqW6WN4/7BkjWuJxLZ0AMOODfjWz+2nzSh9n3SVUD+yw8vJJh75V5jrv4uo QVz+Ksal+fi0gpZJmNv4PO0wrEqBy37OluFag= MIME-Version: 1.0 Received: by 10.229.36.14 with SMTP id r14mr1268277qcd.23.1319013299750; Wed, 19 Oct 2011 01:34:59 -0700 (PDT) Sender: arjuna.sathiaseelan@gmail.com Received: by 10.229.224.69 with HTTP; Wed, 19 Oct 2011 01:34:59 -0700 (PDT) In-Reply-To: References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> <4E932F77.5060006@isi.edu> <4E993917.5030508@erg.abdn.ac.uk> <4E9DD306.4020402@isi.edu> <4E9E7BA0.8060406@erg.abdn.ac.uk> Date: Wed, 19 Oct 2011 09:34:59 +0100 X-Google-Sender-Auth: zqC7YdBuOoTQyrJ3GfF0fr0nqo8 Message-ID: From: Arjuna Sathiaseelan To: ledbat@ietf.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Subject: Re: [ledbat] New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 19 Oct 2011 08:35:01 -0000 I agree to these points too that LEDBAT MUST NOT evolve its initial window to a larger value. To solve the issue of bursts as Joe pointed out, do we need LEDBAT to be rate paced (atleast during the initial phases)? Arjuna > On 19 October 2011 08:26, Gorry Fairhurst wrote: >> On 18/10/2011 20:27, Joe Touch wrote: >>> >>> >>> On 10/18/2011 8:29 AM, David Ros wrote: >>> ... >>>> >>>> Hi all, >>>> >>>> Sorry if I'm getting this totally wrong, but: is it really >>>> *necessary* >>>> that LEDBAT's INIT_CWND be *smaller* than TCP's? As far as I understan= d, >>>> the point of LEDBAT is to be less-than-best-effort over (relatively) >>>> long timescales, or at least over time scales longer than one (initial= ) >>>> RTT. And just starting up as TCP won't make it *more* aggressive than >>>> TCP. Is this correct?? >>> >>> No, but it could make it a lot like TCP if the offered load is in short >>> bursts. If that's not the intent, then the INIT_CWND needs to be smalle= r >>> than TCP's. >>> >>> Joe >>> >>> >> >> I think INIT_CWND should not be significantly bigger than *deployed* TCP >> INIT_CWND. (It may of course be desirable to be smaller or the same, and >> that would benefit in the way Joe suggested). >> >> When I suggested "4" could be OK, this was only 1 larger than current us= age >> for a 1500B MTU, and equivalent for some smaller MTU. That to me is not >> "significantly bigger". It would seem OK, because if this induced conges= tion >> LEDBAT would react within INIT_CWND segments in a conservative way. >> >> If we care about LEDBAT being conservative compared to other TCP session= s, >> then I really think we should not track future new proposals to raise >> INIT_CWND. I suggest this would have side effects: >> >> - It could make LEDBAT more aggressive than *deployed* TCP implementatio= ns, >> that I think would be bad. >> >> - It may require LEDBAT to implement additional algorithms to ensure it = is >> conservative when the larger INIT_CWND induces congestion. Addressing th= is >> would likely add complexity to LEDBAT and make it dependent on these TCP >> updates (if any). >> >> I think a larger INIT_CWND (e.g. by tracking any evolution of TCP's >> INIT_CWND) is unwarranted if the goal is for LEDBAT to target =A0bulk >> less-than-best-effort use. I cannot see the case yet for a LEDBAT INIT-C= WND >> beyond 4. Raising this could indeed save an RTT at the start of a LEDBAT >> flow, but this would be at the expense of making it (slightly) more >> aggressive than presently deployed TCP. This seems undesirable. >> >> I also suggested that MIN_CWND should similarly be fixed to a small numb= er. >> >> Gorry >> >> _______________________________________________ >> ledbat mailing list >> ledbat@ietf.org >> https://www.ietf.org/mailman/listinfo/ledbat >> > > > > -- > http://about.me/arjuna.sathiaseelan > From touch@isi.edu Wed Oct 19 10:16:17 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2320521F8B8A for ; Wed, 19 Oct 2011 10:16:17 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -103.599 X-Spam-Level: X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[AWL=-1.000, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BVJUQkazEjhR for ; Wed, 19 Oct 2011 10:16:16 -0700 (PDT) Received: from darkstar.isi.edu (darkstar.isi.edu [128.9.128.127]) by ietfa.amsl.com (Postfix) with ESMTP id 9444F21F8B73 for ; Wed, 19 Oct 2011 10:16:16 -0700 (PDT) Received: from [172.16.38.64] (205.158.58.226.ptr.us.xo.net [205.158.58.226]) (authenticated bits=0) by darkstar.isi.edu (8.13.8/8.13.8) with ESMTP id p9JHEru5014863 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Wed, 19 Oct 2011 10:14:57 -0700 (PDT) Message-ID: <4E9F058D.4040909@isi.edu> Date: Wed, 19 Oct 2011 10:14:53 -0700 From: Joe Touch User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1 MIME-Version: 1.0 To: gorry@erg.abdn.ac.uk References: <4E92338F.1030601@fandm.edu> <4E93032B.7090006@isi.edu> <201110101846.26115.mirja.kuehlewind@ikr.uni-stuttgart.de> <4E932F77.5060006@isi.edu> <4E993917.5030508@erg.abdn.ac.uk> <4E9DD306.4020402@isi.edu> <4E9E7BA0.8060406@erg.abdn.ac.uk> In-Reply-To: <4E9E7BA0.8060406@erg.abdn.ac.uk> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: Janardhan Iyengar , ledbat@ietf.org Subject: Re: [ledbat] New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 19 Oct 2011 17:16:17 -0000 Hi, Gorry, I agree with your logic, but it suggests INIT_CWND and MIN_CWND should be 2 - those are both clearly safe, clearly smaller than the current deployed base, and would not increase to track TCP's increases. Joe On 10/19/2011 12:26 AM, Gorry Fairhurst wrote: > On 18/10/2011 20:27, Joe Touch wrote: >> >> >> On 10/18/2011 8:29 AM, David Ros wrote: >> ... >>> Hi all, >>> >>> Sorry if I'm getting this totally wrong, but: is it really >>> *necessary* >>> that LEDBAT's INIT_CWND be *smaller* than TCP's? As far as I understand, >>> the point of LEDBAT is to be less-than-best-effort over (relatively) >>> long timescales, or at least over time scales longer than one (initial) >>> RTT. And just starting up as TCP won't make it *more* aggressive than >>> TCP. Is this correct?? >> >> No, but it could make it a lot like TCP if the offered load is in short >> bursts. If that's not the intent, then the INIT_CWND needs to be smaller >> than TCP's. >> >> Joe >> >> > > I think INIT_CWND should not be significantly bigger than *deployed* TCP > INIT_CWND. (It may of course be desirable to be smaller or the same, and > that would benefit in the way Joe suggested). > > When I suggested "4" could be OK, this was only 1 larger than current > usage for a 1500B MTU, and equivalent for some smaller MTU. That to me > is not "significantly bigger". It would seem OK, because if this induced > congestion LEDBAT would react within INIT_CWND segments in a > conservative way. > > If we care about LEDBAT being conservative compared to other TCP > sessions, then I really think we should not track future new proposals > to raise INIT_CWND. I suggest this would have side effects: > > - It could make LEDBAT more aggressive than *deployed* TCP > implementations, that I think would be bad. > > - It may require LEDBAT to implement additional algorithms to ensure it > is conservative when the larger INIT_CWND induces congestion. Addressing > this would likely add complexity to LEDBAT and make it dependent on > these TCP updates (if any). > > I think a larger INIT_CWND (e.g. by tracking any evolution of TCP's > INIT_CWND) is unwarranted if the goal is for LEDBAT to target bulk > less-than-best-effort use. I cannot see the case yet for a LEDBAT > INIT-CWND beyond 4. Raising this could indeed save an RTT at the start > of a LEDBAT flow, but this would be at the expense of making it > (slightly) more aggressive than presently deployed TCP. This seems > undesirable. > > I also suggested that MIN_CWND should similarly be fixed to a small number. > > Gorry From ingemar.s.johansson@ericsson.com Wed Oct 19 22:21:21 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8151E11E807F for ; Wed, 19 Oct 2011 22:21:21 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.599 X-Spam-Level: X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Eq7Wb5eTkhvx for ; Wed, 19 Oct 2011 22:21:20 -0700 (PDT) Received: from mailgw9.se.ericsson.net (mailgw9.se.ericsson.net [193.180.251.57]) by ietfa.amsl.com (Postfix) with ESMTP id 8123721F84CD for ; Wed, 19 Oct 2011 22:21:20 -0700 (PDT) X-AuditID: c1b4fb39-b7bfdae000005125-19-4e9fafce6f66 Received: from esessmw0256.eemea.ericsson.se (Unknown_Domain [153.88.253.125]) by mailgw9.se.ericsson.net (Symantec Mail Security) with SMTP id 25.78.20773.ECFAF9E4; Thu, 20 Oct 2011 07:21:18 +0200 (CEST) Received: from ESESSCMS0366.eemea.ericsson.se ([169.254.1.23]) by esessmw0256.eemea.ericsson.se ([10.2.3.125]) with mapi; Thu, 20 Oct 2011 07:21:18 +0200 From: Ingemar Johansson S To: "ledbat@ietf.org" Date: Thu, 20 Oct 2011 07:21:16 +0200 Thread-Topic: INIT_CWND and MIN_CWND (WAS Re: [ledbat] New Version Notification for draft-ietf-ledbat-congestion-08.txt) Thread-Index: AcyOkYI02ozFAXosSVCtXZH/gpbVWgAVUokA Message-ID: References: In-Reply-To: Accept-Language: sv-SE, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: sv-SE, en-US Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: AAAAAA== Cc: "jana.iyengar@gmail.com" Subject: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 05:21:21 -0000 Hi Change of subject name.=20 I have no strong opinion on this but I would like to ask how likely it is t= hat LEDBAT is used for shortlived flows in the first place. The CC-algo doe= s not seem to me as the obvious pick of I was looking for instance for fast= HTTP request/response times. Given that it seems like minor issue what the= INIT_CWND is if the flows are 10minutes in duration or perhaps more. MIN_CWND is another matter, that should possibly be fixed to relatively sma= ll value like 2. /Ingemar >=20 > Message: 3 > Date: Wed, 19 Oct 2011 10:14:53 -0700 > From: Joe Touch > To: gorry@erg.abdn.ac.uk > Cc: Janardhan Iyengar , ledbat@ietf.org > Subject: Re: [ledbat] New Version Notification for > draft-ietf-ledbat-congestion-08.txt > Message-ID: <4E9F058D.4040909@isi.edu> > Content-Type: text/plain; charset=3DISO-8859-1; format=3Dflowed >=20 > Hi, Gorry, >=20 > I agree with your logic, but it suggests INIT_CWND and=20 > MIN_CWND should be 2 - those are both clearly safe, clearly=20 > smaller than the current deployed base, and would not=20 > increase to track TCP's increases. >=20 > Joe >=20 > On 10/19/2011 12:26 AM, Gorry Fairhurst wrote: > > On 18/10/2011 20:27, Joe Touch wrote: > >> > >> > >> On 10/18/2011 8:29 AM, David Ros wrote: > >> ... > >>> Hi all, > >>> > >>> Sorry if I'm getting this totally wrong, but: is it really > >>> *necessary* > >>> that LEDBAT's INIT_CWND be *smaller* than TCP's? As far as I=20 > >>> understand, the point of LEDBAT is to be=20 > less-than-best-effort over=20 > >>> (relatively) long timescales, or at least over time scales longer=20 > >>> than one (initial) RTT. And just starting up as TCP won't make it=20 > >>> *more* aggressive than TCP. Is this correct?? > >> > >> No, but it could make it a lot like TCP if the offered load is in=20 > >> short bursts. If that's not the intent, then the INIT_CWND=20 > needs to=20 > >> be smaller than TCP's. > >> > >> Joe > >> > >> > > > > I think INIT_CWND should not be significantly bigger than=20 > *deployed*=20 > > TCP INIT_CWND. (It may of course be desirable to be smaller or the=20 > > same, and that would benefit in the way Joe suggested). > > > > When I suggested "4" could be OK, this was only 1 larger=20 > than current=20 > > usage for a 1500B MTU, and equivalent for some smaller MTU.=20 > That to me=20 > > is not "significantly bigger". It would seem OK, because if this=20 > > induced congestion LEDBAT would react within INIT_CWND=20 > segments in a=20 > > conservative way. > > > > If we care about LEDBAT being conservative compared to other TCP=20 > > sessions, then I really think we should not track future=20 > new proposals=20 > > to raise INIT_CWND. I suggest this would have side effects: > > > > - It could make LEDBAT more aggressive than *deployed* TCP=20 > > implementations, that I think would be bad. > > > > - It may require LEDBAT to implement additional algorithms=20 > to ensure=20 > > it is conservative when the larger INIT_CWND induces congestion.=20 > > Addressing this would likely add complexity to LEDBAT and make it=20 > > dependent on these TCP updates (if any). > > > > I think a larger INIT_CWND (e.g. by tracking any evolution of TCP's > > INIT_CWND) is unwarranted if the goal is for LEDBAT to target bulk=20 > > less-than-best-effort use. I cannot see the case yet for a LEDBAT=20 > > INIT-CWND beyond 4. Raising this could indeed save an RTT=20 > at the start=20 > > of a LEDBAT flow, but this would be at the expense of making it > > (slightly) more aggressive than presently deployed TCP. This seems=20 > > undesirable. > > > > I also suggested that MIN_CWND should similarly be fixed to=20 > a small number. > > > > Gorry >=20 >=20 > ------------------------------ >=20 > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat >=20 >=20 > End of ledbat Digest, Vol 33, Issue 9 > ************************************* > = From weigengyu@vip.sina.com Thu Oct 20 00:20:02 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 798A111E8085 for ; Thu, 20 Oct 2011 00:20:02 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -0.293 X-Spam-Level: X-Spam-Status: No, score=-0.293 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HELO_MISMATCH_COM=0.553, MIME_BASE64_TEXT=1.753] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UZgmY1ftcpYL for ; Thu, 20 Oct 2011 00:20:01 -0700 (PDT) Received: from smtp-6-98.vip.sina.com (mail3-173.sinamail.sina.com.cn [202.108.3.173]) by ietfa.amsl.com (Postfix) with ESMTP id 6492711E80A1 for ; Thu, 20 Oct 2011 00:19:57 -0700 (PDT) Received: from irsmtp-xd5-211.sinamail.sina.com.cn (unknown [10.55.5.211]) by smtp-6-98.vip.sina.com (SINAMAIL) with ESMTP id 848951DDC43 for ; Thu, 20 Oct 2011 15:19:52 +0800 (CST) X-Originating-IP: [59.64.159.158] X-Auth-ID: weigengyu X-Originating-IP: 59.64.159.158 Received: from unknown (HELO Notebook) ([59.64.159.158]) by irsmtp-xd5-211.sinamail.sina.com.cn with ESMTP; 20 Oct 2011 15:19:51 +0800 Message-ID: <606428E53EC046A09978B6CD863F7559@Notebook> From: "Gengyu WEI" To: "Ingemar Johansson S" References: Date: Thu, 20 Oct 2011 15:19:49 +0800 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: base64 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.5931 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.6109 Cc: ledbat@ietf.org Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 07:20:02 -0000 SW5nZW1hciwgDQoNCkkgYWdyZWUuIA0KDQpBbmQgaWYgaGUgc2hvcnRsaXZlZCBmbG93cyBhbmQg c21hbGwgYnVsayBvZiBkYXRhIGFyZSBzdXBwb3NlZCwgDQppdCBpcyBzaWduaWZpY2FudCB0byBp bmNyZWFzZSB0aGUgd2luZG93IGV4cG9uZW50aWFsbHk/ICANCg0KR2VuZ3l1DQoNCi0tLS0tIE9y aWdpbmFsIE1lc3NhZ2UgLS0tLS0gDQpGcm9tOiAiSW5nZW1hciBKb2hhbnNzb24gUyIgPGluZ2Vt YXIucy5qb2hhbnNzb25AZXJpY3Nzb24uY29tPg0KVG86IDxsZWRiYXRAaWV0Zi5vcmc+DQpDYzog PGphbmEuaXllbmdhckBnbWFpbC5jb20+DQpTZW50OiBUaHVyc2RheSwgT2N0b2JlciAyMCwgMjAx MSAxOjIxIFBNDQpTdWJqZWN0OiBbbGVkYmF0XSBJTklUX0NXTkQgYW5kIE1JTl9DV05EIChXQVMg UmU6IE5ldyBWZXJzaW9uIE5vdGlmaWNhdGlvbiBmb3IgZHJhZnQtaWV0Zi1sZWRiYXQtY29uZ2Vz dGlvbi0wOC50eHQpDQoNCg0KPiBIaQ0KPiANCj4gQ2hhbmdlIG9mIHN1YmplY3QgbmFtZS4gDQo+ IEkgaGF2ZSBubyBzdHJvbmcgb3BpbmlvbiBvbiB0aGlzIGJ1dCBJIHdvdWxkIGxpa2UgdG8gYXNr IGhvdyBsaWtlbHkgaXQgaXMgdGhhdCBMRURCQVQgaXMgdXNlZCBmb3Igc2hvcnRsaXZlZCBmbG93 cyBpbiB0aGUgZmlyc3QgcGxhY2UuIFRoZSBDQy1hbGdvIGRvZXMgbm90IHNlZW0gdG8gbWUgYXMg dGhlIG9idmlvdXMgcGljayBvZiBJIHdhcyBsb29raW5nIGZvciBpbnN0YW5jZSBmb3IgZmFzdCBI VFRQIHJlcXVlc3QvcmVzcG9uc2UgdGltZXMuIEdpdmVuIHRoYXQgaXQgc2VlbXMgbGlrZSBtaW5v ciBpc3N1ZSB3aGF0IHRoZSBJTklUX0NXTkQgaXMgaWYgdGhlIGZsb3dzIGFyZSAxMG1pbnV0ZXMg aW4gZHVyYXRpb24gb3IgcGVyaGFwcyBtb3JlLg0KPiBNSU5fQ1dORCBpcyBhbm90aGVyIG1hdHRl ciwgdGhhdCBzaG91bGQgcG9zc2libHkgYmUgZml4ZWQgdG8gcmVsYXRpdmVseSBzbWFsbCB2YWx1 ZSBsaWtlIDIuDQo+IA0KPiAvSW5nZW1hcg0KPiANCj4+IA0KPj4gTWVzc2FnZTogMw0KPj4gRGF0 ZTogV2VkLCAxOSBPY3QgMjAxMSAxMDoxNDo1MyAtMDcwMA0KPj4gRnJvbTogSm9lIFRvdWNoIDx0 b3VjaEBpc2kuZWR1Pg0KPj4gVG86IGdvcnJ5QGVyZy5hYmRuLmFjLnVrDQo+PiBDYzogSmFuYXJk aGFuIEl5ZW5nYXIgPGphbmEuaXllbmdhckBnbWFpbC5jb20+LCBsZWRiYXRAaWV0Zi5vcmcNCj4+ IFN1YmplY3Q6IFJlOiBbbGVkYmF0XSBOZXcgVmVyc2lvbiBOb3RpZmljYXRpb24gZm9yDQo+PiBk cmFmdC1pZXRmLWxlZGJhdC1jb25nZXN0aW9uLTA4LnR4dA0KPj4gTWVzc2FnZS1JRDogPDRFOUYw NThELjQwNDA5MDlAaXNpLmVkdT4NCj4+IENvbnRlbnQtVHlwZTogdGV4dC9wbGFpbjsgY2hhcnNl dD1JU08tODg1OS0xOyBmb3JtYXQ9Zmxvd2VkDQo+PiANCj4+IEhpLCBHb3JyeSwNCj4+IA0KPj4g SSBhZ3JlZSB3aXRoIHlvdXIgbG9naWMsIGJ1dCBpdCBzdWdnZXN0cyBJTklUX0NXTkQgYW5kIA0K Pj4gTUlOX0NXTkQgc2hvdWxkIGJlIDIgLSB0aG9zZSBhcmUgYm90aCBjbGVhcmx5IHNhZmUsIGNs ZWFybHkgDQo+PiBzbWFsbGVyIHRoYW4gdGhlIGN1cnJlbnQgZGVwbG95ZWQgYmFzZSwgYW5kIHdv dWxkIG5vdCANCj4+IGluY3JlYXNlIHRvIHRyYWNrIFRDUCdzIGluY3JlYXNlcy4NCj4+IA0KPj4g Sm9lDQo+PiANCj4+IE9uIDEwLzE5LzIwMTEgMTI6MjYgQU0sIEdvcnJ5IEZhaXJodXJzdCB3cm90 ZToNCj4+ID4gT24gMTgvMTAvMjAxMSAyMDoyNywgSm9lIFRvdWNoIHdyb3RlOg0KPj4gPj4NCj4+ ID4+DQo+PiA+PiBPbiAxMC8xOC8yMDExIDg6MjkgQU0sIERhdmlkIFJvcyB3cm90ZToNCj4+ID4+ IC4uLg0KPj4gPj4+IEhpIGFsbCwNCj4+ID4+Pg0KPj4gPj4+IFNvcnJ5IGlmIEknbSBnZXR0aW5n IHRoaXMgdG90YWxseSB3cm9uZywgYnV0OiBpcyBpdCByZWFsbHkNCj4+ID4+PiAqbmVjZXNzYXJ5 Kg0KPj4gPj4+IHRoYXQgTEVEQkFUJ3MgSU5JVF9DV05EIGJlICpzbWFsbGVyKiB0aGFuIFRDUCdz PyBBcyBmYXIgYXMgSSANCj4+ID4+PiB1bmRlcnN0YW5kLCB0aGUgcG9pbnQgb2YgTEVEQkFUIGlz IHRvIGJlIA0KPj4gbGVzcy10aGFuLWJlc3QtZWZmb3J0IG92ZXIgDQo+PiA+Pj4gKHJlbGF0aXZl bHkpIGxvbmcgdGltZXNjYWxlcywgb3IgYXQgbGVhc3Qgb3ZlciB0aW1lIHNjYWxlcyBsb25nZXIg DQo+PiA+Pj4gdGhhbiBvbmUgKGluaXRpYWwpIFJUVC4gQW5kIGp1c3Qgc3RhcnRpbmcgdXAgYXMg VENQIHdvbid0IG1ha2UgaXQgDQo+PiA+Pj4gKm1vcmUqIGFnZ3Jlc3NpdmUgdGhhbiBUQ1AuIElz IHRoaXMgY29ycmVjdD8/DQo+PiA+Pg0KPj4gPj4gTm8sIGJ1dCBpdCBjb3VsZCBtYWtlIGl0IGEg bG90IGxpa2UgVENQIGlmIHRoZSBvZmZlcmVkIGxvYWQgaXMgaW4gDQo+PiA+PiBzaG9ydCBidXJz dHMuIElmIHRoYXQncyBub3QgdGhlIGludGVudCwgdGhlbiB0aGUgSU5JVF9DV05EIA0KPj4gbmVl ZHMgdG8gDQo+PiA+PiBiZSBzbWFsbGVyIHRoYW4gVENQJ3MuDQo+PiA+Pg0KPj4gPj4gSm9lDQo+ PiA+Pg0KPj4gPj4NCj4+ID4NCj4+ID4gSSB0aGluayBJTklUX0NXTkQgc2hvdWxkIG5vdCBiZSBz aWduaWZpY2FudGx5IGJpZ2dlciB0aGFuIA0KPj4gKmRlcGxveWVkKiANCj4+ID4gVENQIElOSVRf Q1dORC4gKEl0IG1heSBvZiBjb3Vyc2UgYmUgZGVzaXJhYmxlIHRvIGJlIHNtYWxsZXIgb3IgdGhl IA0KPj4gPiBzYW1lLCBhbmQgdGhhdCB3b3VsZCBiZW5lZml0IGluIHRoZSB3YXkgSm9lIHN1Z2dl c3RlZCkuDQo+PiA+DQo+PiA+IFdoZW4gSSBzdWdnZXN0ZWQgIjQiIGNvdWxkIGJlIE9LLCB0aGlz IHdhcyBvbmx5IDEgbGFyZ2VyIA0KPj4gdGhhbiBjdXJyZW50IA0KPj4gPiB1c2FnZSBmb3IgYSAx NTAwQiBNVFUsIGFuZCBlcXVpdmFsZW50IGZvciBzb21lIHNtYWxsZXIgTVRVLiANCj4+IFRoYXQg dG8gbWUgDQo+PiA+IGlzIG5vdCAic2lnbmlmaWNhbnRseSBiaWdnZXIiLiBJdCB3b3VsZCBzZWVt IE9LLCBiZWNhdXNlIGlmIHRoaXMgDQo+PiA+IGluZHVjZWQgY29uZ2VzdGlvbiBMRURCQVQgd291 bGQgcmVhY3Qgd2l0aGluIElOSVRfQ1dORCANCj4+IHNlZ21lbnRzIGluIGEgDQo+PiA+IGNvbnNl cnZhdGl2ZSB3YXkuDQo+PiA+DQo+PiA+IElmIHdlIGNhcmUgYWJvdXQgTEVEQkFUIGJlaW5nIGNv bnNlcnZhdGl2ZSBjb21wYXJlZCB0byBvdGhlciBUQ1AgDQo+PiA+IHNlc3Npb25zLCB0aGVuIEkg cmVhbGx5IHRoaW5rIHdlIHNob3VsZCBub3QgdHJhY2sgZnV0dXJlIA0KPj4gbmV3IHByb3Bvc2Fs cyANCj4+ID4gdG8gcmFpc2UgSU5JVF9DV05ELiBJIHN1Z2dlc3QgdGhpcyB3b3VsZCBoYXZlIHNp ZGUgZWZmZWN0czoNCj4+ID4NCj4+ID4gLSBJdCBjb3VsZCBtYWtlIExFREJBVCBtb3JlIGFnZ3Jl c3NpdmUgdGhhbiAqZGVwbG95ZWQqIFRDUCANCj4+ID4gaW1wbGVtZW50YXRpb25zLCB0aGF0IEkg dGhpbmsgd291bGQgYmUgYmFkLg0KPj4gPg0KPj4gPiAtIEl0IG1heSByZXF1aXJlIExFREJBVCB0 byBpbXBsZW1lbnQgYWRkaXRpb25hbCBhbGdvcml0aG1zIA0KPj4gdG8gZW5zdXJlIA0KPj4gPiBp dCBpcyBjb25zZXJ2YXRpdmUgd2hlbiB0aGUgbGFyZ2VyIElOSVRfQ1dORCBpbmR1Y2VzIGNvbmdl c3Rpb24uIA0KPj4gPiBBZGRyZXNzaW5nIHRoaXMgd291bGQgbGlrZWx5IGFkZCBjb21wbGV4aXR5 IHRvIExFREJBVCBhbmQgbWFrZSBpdCANCj4+ID4gZGVwZW5kZW50IG9uIHRoZXNlIFRDUCB1cGRh dGVzIChpZiBhbnkpLg0KPj4gPg0KPj4gPiBJIHRoaW5rIGEgbGFyZ2VyIElOSVRfQ1dORCAoZS5n LiBieSB0cmFja2luZyBhbnkgZXZvbHV0aW9uIG9mIFRDUCdzDQo+PiA+IElOSVRfQ1dORCkgaXMg dW53YXJyYW50ZWQgaWYgdGhlIGdvYWwgaXMgZm9yIExFREJBVCB0byB0YXJnZXQgYnVsayANCj4+ ID4gbGVzcy10aGFuLWJlc3QtZWZmb3J0IHVzZS4gSSBjYW5ub3Qgc2VlIHRoZSBjYXNlIHlldCBm b3IgYSBMRURCQVQgDQo+PiA+IElOSVQtQ1dORCBiZXlvbmQgNC4gUmFpc2luZyB0aGlzIGNvdWxk IGluZGVlZCBzYXZlIGFuIFJUVCANCj4+IGF0IHRoZSBzdGFydCANCj4+ID4gb2YgYSBMRURCQVQg ZmxvdywgYnV0IHRoaXMgd291bGQgYmUgYXQgdGhlIGV4cGVuc2Ugb2YgbWFraW5nIGl0DQo+PiA+ IChzbGlnaHRseSkgbW9yZSBhZ2dyZXNzaXZlIHRoYW4gcHJlc2VudGx5IGRlcGxveWVkIFRDUC4g VGhpcyBzZWVtcyANCj4+ID4gdW5kZXNpcmFibGUuDQo+PiA+DQo+PiA+IEkgYWxzbyBzdWdnZXN0 ZWQgdGhhdCBNSU5fQ1dORCBzaG91bGQgc2ltaWxhcmx5IGJlIGZpeGVkIHRvIA0KPj4gYSBzbWFs bCBudW1iZXIuDQo+PiA+DQo+PiA+IEdvcnJ5DQo+PiANCj4+IA0KPj4gLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tDQo+PiANCj4+IF9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fDQo+PiBsZWRiYXQgbWFpbGluZyBsaXN0DQo+PiBsZWRiYXRAaWV0Zi5v cmcNCj4+IGh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8vbGVkYmF0DQo+PiAN Cj4+IA0KPj4gRW5kIG9mIGxlZGJhdCBEaWdlc3QsIFZvbCAzMywgSXNzdWUgOQ0KPj4gKioqKioq KioqKioqKioqKioqKioqKioqKioqKioqKioqKioqKg0KPj4gDQo+IF9fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fDQo+IGxlZGJhdCBtYWlsaW5nIGxpc3QNCj4g bGVkYmF0QGlldGYub3JnDQo+IGh0dHBzOi8vd3d3LmlldGYub3JnL21haWxtYW4vbGlzdGluZm8v bGVkYmF0DQo+ From David.Ros@telecom-bretagne.eu Thu Oct 20 01:35:54 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A170321F84F8 for ; Thu, 20 Oct 2011 01:35:54 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.949 X-Spam-Level: X-Spam-Status: No, score=-1.949 tagged_above=-999 required=5 tests=[AWL=0.300, BAYES_00=-2.599, HELO_EQ_FR=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id olI1gRVEtlIV for ; Thu, 20 Oct 2011 01:35:54 -0700 (PDT) Received: from coliposte.enst-bretagne.fr (coliposte.enst-bretagne.fr [192.108.115.12]) by ietfa.amsl.com (Postfix) with ESMTP id 7A7E221F84DF for ; Thu, 20 Oct 2011 01:35:53 -0700 (PDT) Received: from localhost (localhost.localdomain [127.0.0.1]) by coliposte.enst-bretagne.fr (8.13.7/8.13.7/2009.11.10) with ESMTP id p9K8Zjnh025281; Thu, 20 Oct 2011 10:35:45 +0200 Received: from courrier.enst-bretagne.fr (smtps.enst-bretagne.fr [10.29.90.4]) by coliposte.enst-bretagne.fr (8.13.7/8.13.7/2009.11.10) with ESMTP id p9K8Zbc6025238; Thu, 20 Oct 2011 10:35:41 +0200 Received: from eduroam-193-157-113-173.uio.no (passerelle-interne.enst-bretagne.fr [192.108.117.210]) (user=dros mech=PLAIN bits=0) by courrier.enst-bretagne.fr (8.13.8/8.13.8/2010.02.22) with ESMTP id p9K8ZYPs017002; Thu, 20 Oct 2011 10:35:35 +0200 Mime-Version: 1.0 (Apple Message framework v1084) Content-Type: text/plain; charset=iso-8859-1 From: David Ros In-Reply-To: Date: Thu, 20 Oct 2011 10:35:34 +0200 Content-Transfer-Encoding: quoted-printable Message-Id: <52C453CB-2B0D-4B74-AB8D-17DC524B8AF7@telecom-bretagne.eu> References: To: Ingemar Johansson S X-Mailer: Apple Mail (2.1084) X-Virus-Scanned: amavisd-new at enst-bretagne.fr Cc: "jana.iyengar@gmail.com" , "ledbat@ietf.org" Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 08:35:54 -0000 Hi, INIT_CWND =3D MIN_CWND =3D 2 sounds reasonable, and it seems to take = care of the points raised -- including apps that generate nothing but = short flows using LEDBAT (seems like a bad choice of CC for such apps, = IMHO) -- except one. The only thing remaining would be the filtering of delay values (the = draft says INIT_CWND =3D 4 to initialize the filter in the first RTT), = but I'd venture that this is not necessary? i.e. taking a bit more time = to seed the filter doesn't look as if it'll break LEDBAT's CC? Thanks, David. Le 20 oct. 2011 =E0 07:21, Ingemar Johansson S a =E9crit : > Hi >=20 > Change of subject name.=20 > I have no strong opinion on this but I would like to ask how likely it = is that LEDBAT is used for shortlived flows in the first place. The = CC-algo does not seem to me as the obvious pick of I was looking for = instance for fast HTTP request/response times. Given that it seems like = minor issue what the INIT_CWND is if the flows are 10minutes in duration = or perhaps more. > MIN_CWND is another matter, that should possibly be fixed to = relatively small value like 2. >=20 > /Ingemar >=20 >>=20 >> Message: 3 >> Date: Wed, 19 Oct 2011 10:14:53 -0700 >> From: Joe Touch >> To: gorry@erg.abdn.ac.uk >> Cc: Janardhan Iyengar , ledbat@ietf.org >> Subject: Re: [ledbat] New Version Notification for >> draft-ietf-ledbat-congestion-08.txt >> Message-ID: <4E9F058D.4040909@isi.edu> >> Content-Type: text/plain; charset=3DISO-8859-1; format=3Dflowed >>=20 >> Hi, Gorry, >>=20 >> I agree with your logic, but it suggests INIT_CWND and=20 >> MIN_CWND should be 2 - those are both clearly safe, clearly=20 >> smaller than the current deployed base, and would not=20 >> increase to track TCP's increases. >>=20 >> Joe >>=20 >> On 10/19/2011 12:26 AM, Gorry Fairhurst wrote: >>> On 18/10/2011 20:27, Joe Touch wrote: >>>>=20 >>>>=20 >>>> On 10/18/2011 8:29 AM, David Ros wrote: >>>> ... >>>>> Hi all, >>>>>=20 >>>>> Sorry if I'm getting this totally wrong, but: is it really >>>>> *necessary* >>>>> that LEDBAT's INIT_CWND be *smaller* than TCP's? As far as I=20 >>>>> understand, the point of LEDBAT is to be=20 >> less-than-best-effort over=20 >>>>> (relatively) long timescales, or at least over time scales longer=20= >>>>> than one (initial) RTT. And just starting up as TCP won't make it=20= >>>>> *more* aggressive than TCP. Is this correct?? >>>>=20 >>>> No, but it could make it a lot like TCP if the offered load is in=20= >>>> short bursts. If that's not the intent, then the INIT_CWND=20 >> needs to=20 >>>> be smaller than TCP's. >>>>=20 >>>> Joe >>>>=20 >>>>=20 >>>=20 >>> I think INIT_CWND should not be significantly bigger than=20 >> *deployed*=20 >>> TCP INIT_CWND. (It may of course be desirable to be smaller or the=20= >>> same, and that would benefit in the way Joe suggested). >>>=20 >>> When I suggested "4" could be OK, this was only 1 larger=20 >> than current=20 >>> usage for a 1500B MTU, and equivalent for some smaller MTU.=20 >> That to me=20 >>> is not "significantly bigger". It would seem OK, because if this=20 >>> induced congestion LEDBAT would react within INIT_CWND=20 >> segments in a=20 >>> conservative way. >>>=20 >>> If we care about LEDBAT being conservative compared to other TCP=20 >>> sessions, then I really think we should not track future=20 >> new proposals=20 >>> to raise INIT_CWND. I suggest this would have side effects: >>>=20 >>> - It could make LEDBAT more aggressive than *deployed* TCP=20 >>> implementations, that I think would be bad. >>>=20 >>> - It may require LEDBAT to implement additional algorithms=20 >> to ensure=20 >>> it is conservative when the larger INIT_CWND induces congestion.=20 >>> Addressing this would likely add complexity to LEDBAT and make it=20 >>> dependent on these TCP updates (if any). >>>=20 >>> I think a larger INIT_CWND (e.g. by tracking any evolution of TCP's >>> INIT_CWND) is unwarranted if the goal is for LEDBAT to target bulk=20= >>> less-than-best-effort use. I cannot see the case yet for a LEDBAT=20 >>> INIT-CWND beyond 4. Raising this could indeed save an RTT=20 >> at the start=20 >>> of a LEDBAT flow, but this would be at the expense of making it >>> (slightly) more aggressive than presently deployed TCP. This seems=20= >>> undesirable. >>>=20 >>> I also suggested that MIN_CWND should similarly be fixed to=20 >> a small number. >>>=20 >>> Gorry >>=20 >>=20 >> ------------------------------ >>=20 >> _______________________________________________ >> ledbat mailing list >> ledbat@ietf.org >> https://www.ietf.org/mailman/listinfo/ledbat >>=20 >>=20 >> End of ledbat Digest, Vol 33, Issue 9 >> ************************************* >>=20 > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat >=20 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D David ROS http://www.rennes.enst-bretagne.fr/~dros/ "It would seem that you have no useful skill or talent whatsoever," he = said. "Have you thought of going into teaching?" -- Terry Pratchett From David.Ros@telecom-bretagne.eu Thu Oct 20 02:11:08 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 821D221F84F8 for ; Thu, 20 Oct 2011 02:11:08 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.099 X-Spam-Level: X-Spam-Status: No, score=-2.099 tagged_above=-999 required=5 tests=[AWL=0.150, BAYES_00=-2.599, HELO_EQ_FR=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pNbShu-p5JfI for ; Thu, 20 Oct 2011 02:11:07 -0700 (PDT) Received: from coliposte.enst-bretagne.fr (coliposte.enst-bretagne.fr [192.108.115.12]) by ietfa.amsl.com (Postfix) with ESMTP id 551E821F84DF for ; Thu, 20 Oct 2011 02:11:07 -0700 (PDT) Received: from localhost (localhost.localdomain [127.0.0.1]) by coliposte.enst-bretagne.fr (8.13.7/8.13.7/2009.11.10) with ESMTP id p9K9B2n5008638 for ; Thu, 20 Oct 2011 11:11:03 +0200 Received: from courrier.enst-bretagne.fr (smtps.enst-bretagne.fr [10.29.90.4]) by coliposte.enst-bretagne.fr (8.13.7/8.13.7/2009.11.10) with ESMTP id p9K9AsqU008550 for ; Thu, 20 Oct 2011 11:10:57 +0200 Received: from eduroam-193-157-113-173.uio.no (passerelle-interne.enst-bretagne.fr [192.108.117.210]) (user=dros mech=PLAIN bits=0) by courrier.enst-bretagne.fr (8.13.8/8.13.8/2010.02.22) with ESMTP id p9K9Ap7F000532 for ; Thu, 20 Oct 2011 11:10:51 +0200 Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Apple Message framework v1084) From: David Ros In-Reply-To: <791AD3077F94194BB2BDD13565B6295D1D0A463B@DAPHNIS.office.hd> Date: Thu, 20 Oct 2011 11:10:51 +0200 Content-Transfer-Encoding: quoted-printable Message-Id: <32BB54A8-D41C-45C6-949E-F3060A6FF966@telecom-bretagne.eu> References: <791AD3077F94194BB2BDD13565B6295D1D0A463B@DAPHNIS.office.hd> To: ledbat@ietf.org X-Mailer: Apple Mail (2.1084) X-Virus-Scanned: amavisd-new at enst-bretagne.fr Subject: [ledbat] initialization of current_delays in draft-ietf-ledbat-congestion-08 X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 09:11:08 -0000 Hi, After reading the current version of the draft, I was wondering (sorry = if it's a very stupid question): Why does the current_delays list have to be initialized to +INFINITY? = ("on initialization:", sec. 3.4.2) I understand why that's the case for = base_delays, but not for current_delays. Thanks, David. PS: Jana and Mirja, I'll send you offline a list of typos I found. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D David ROS http://www.rennes.enst-bretagne.fr/~dros/ "But we're a university! [...] What sort of people would we be if we = didn't go into the Library?" "Students." -- Terry Pratchett From touch@isi.edu Thu Oct 20 10:30:09 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 552E221F8C55 for ; Thu, 20 Oct 2011 10:30:09 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.949 X-Spam-Level: X-Spam-Status: No, score=-102.949 tagged_above=-999 required=5 tests=[AWL=-0.350, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Fw7dNMS+y7R7 for ; Thu, 20 Oct 2011 10:30:08 -0700 (PDT) Received: from vapor.isi.edu (vapor.isi.edu [128.9.64.64]) by ietfa.amsl.com (Postfix) with ESMTP id 9FCDC21F8C36 for ; Thu, 20 Oct 2011 10:30:08 -0700 (PDT) Received: from [128.9.160.166] (abc.isi.edu [128.9.160.166]) (authenticated bits=0) by vapor.isi.edu (8.13.8/8.13.8) with ESMTP id p9KHTT64015607 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Thu, 20 Oct 2011 10:29:29 -0700 (PDT) Message-ID: <4EA05A79.7080700@isi.edu> Date: Thu, 20 Oct 2011 10:29:29 -0700 From: Joe Touch User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1 MIME-Version: 1.0 To: Ingemar Johansson S References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: "jana.iyengar@gmail.com" , "ledbat@ietf.org" Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 17:30:09 -0000 On 10/19/2011 10:21 PM, Ingemar Johansson S wrote: > Hi > > Change of subject name. > I have no strong opinion on this but I would like to ask how likely > it is that LEDBAT is used for shortlived flows in the first place. INIT_CWND comes into play during any restart, e.g., after a gap in the offered load. While I don't expect LEDBAD would be used primarily for shortlived flows: 1) we can't know that, so should plan accordingly 2) long-lived flows with frequent gaps do exist, and should not cause a problem either Joe From gorry@erg.abdn.ac.uk Thu Oct 20 11:15:24 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1924F21F8B81 for ; Thu, 20 Oct 2011 11:15:24 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.539 X-Spam-Level: X-Spam-Status: No, score=-102.539 tagged_above=-999 required=5 tests=[AWL=0.060, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 9v3JHBp355gv for ; Thu, 20 Oct 2011 11:15:23 -0700 (PDT) Received: from erg.abdn.ac.uk (dee.erg.abdn.ac.uk [IPv6:2001:630:241:204:203:baff:fe9a:8c9b]) by ietfa.amsl.com (Postfix) with ESMTP id 2792521F8B75 for ; Thu, 20 Oct 2011 11:15:22 -0700 (PDT) Received: from gorry-mac.erg.abdn.ac.uk (gorry-mac.erg.abdn.ac.uk [139.133.207.5]) (authenticated bits=0) by erg.abdn.ac.uk (8.13.4/8.13.4) with ESMTP id p9KIFEjf000276 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES128-SHA bits=128 verify=NOT); Thu, 20 Oct 2011 19:15:14 +0100 (BST) Message-ID: <4EA06532.8040204@erg.abdn.ac.uk> Date: Thu, 20 Oct 2011 19:15:14 +0100 From: Gorry Fairhurst Organization: The University of Aberdeen is a charity registered in Scotland, No SC013683. User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1 MIME-Version: 1.0 To: Joe Touch References: <4EA05A79.7080700@isi.edu> In-Reply-To: <4EA05A79.7080700@isi.edu> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ERG-MailScanner: Found to be clean X-ERG-MailScanner-From: gorry@erg.abdn.ac.uk Cc: Ingemar Johansson S , "jana.iyengar@gmail.com" , "ledbat@ietf.org" Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: gorry@erg.abdn.ac.uk List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 18:15:24 -0000 On 20/10/2011 18:29, Joe Touch wrote: > > On 10/19/2011 10:21 PM, Ingemar Johansson S wrote: >> Hi >> >> Change of subject name. >> I have no strong opinion on this but I would like to ask how likely >> it is that LEDBAT is used for shortlived flows in the first place. > > INIT_CWND comes into play during any restart, e.g., after a gap in the > offered load. > Is that the "restart window" (RW) in TCP? > While I don't expect LEDBAD would be used primarily for shortlived flows: > 1) we can't know that, so should plan accordingly > > 2) long-lived flows with frequent gaps do exist, and > should not cause a problem either > I agree with both of these. > Joe > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat > Gorry From David.Ros@telecom-bretagne.eu Thu Oct 20 11:57:20 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 070B021F8B57 for ; Thu, 20 Oct 2011 11:57:20 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.149 X-Spam-Level: X-Spam-Status: No, score=-2.149 tagged_above=-999 required=5 tests=[AWL=0.100, BAYES_00=-2.599, HELO_EQ_FR=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6BwXfQxPC75d for ; Thu, 20 Oct 2011 11:57:19 -0700 (PDT) Received: from coliposte.enst-bretagne.fr (coliposte.enst-bretagne.fr [192.108.115.12]) by ietfa.amsl.com (Postfix) with ESMTP id 1866421F8B4F for ; Thu, 20 Oct 2011 11:57:18 -0700 (PDT) Received: from localhost (localhost.localdomain [127.0.0.1]) by coliposte.enst-bretagne.fr (8.13.7/8.13.7/2009.11.10) with ESMTP id p9KIvGSG015084; Thu, 20 Oct 2011 20:57:16 +0200 Received: from courrier.enst-bretagne.fr (smtps.enst-bretagne.fr [10.29.90.4]) by coliposte.enst-bretagne.fr (8.13.7/8.13.7/2009.11.10) with ESMTP id p9KIvBvi015074; Thu, 20 Oct 2011 20:57:14 +0200 Received: from eduroam-193-157-110-173.uio.no (passerelle-interne.enst-bretagne.fr [192.108.117.210]) (user=dros mech=PLAIN bits=0) by courrier.enst-bretagne.fr (8.13.8/8.13.8/2010.02.22) with ESMTP id p9KIv55W025755; Thu, 20 Oct 2011 20:57:05 +0200 Mime-Version: 1.0 (Apple Message framework v1084) Content-Type: text/plain; charset=iso-8859-1 From: David Ros In-Reply-To: <4EA05A79.7080700@isi.edu> Date: Thu, 20 Oct 2011 20:57:04 +0200 Content-Transfer-Encoding: quoted-printable Message-Id: References: <4EA05A79.7080700@isi.edu> To: Joe Touch X-Mailer: Apple Mail (2.1084) X-Virus-Scanned: amavisd-new at enst-bretagne.fr Cc: Ingemar Johansson S , "jana.iyengar@gmail.com" , "ledbat@ietf.org" Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 18:57:20 -0000 Le 20 oct. 2011 =E0 19:29, Joe Touch a =E9crit : >=20 >=20 > On 10/19/2011 10:21 PM, Ingemar Johansson S wrote: >> Hi >>=20 >> Change of subject name. >> I have no strong opinion on this but I would like to ask how likely >> it is that LEDBAT is used for shortlived flows in the first place. >=20 > INIT_CWND comes into play during any restart, e.g., after a gap in the = offered load. >=20 Isn't that the role of max_allowed_cwnd, instead? (section 3.4.2) This algorithm restricts cwnd growth after a period of inactivity, where the cwnd is clamped down to a little more than flightsize using max_allowed_cwnd Thanks, David. > While I don't expect LEDBAD would be used primarily for shortlived = flows: > 1) we can't know that, so should plan accordingly > 2) long-lived flows with frequent gaps do exist, and > should not cause a problem either >=20 > Joe > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat >=20 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D David ROS http://www.rennes.enst-bretagne.fr/~dros/ In theory, theory and practice are the same. In practice, they are = different. -- Author unknown From touch@isi.edu Thu Oct 20 13:26:23 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 065E211E8090 for ; Thu, 20 Oct 2011 13:26:23 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.832 X-Spam-Level: X-Spam-Status: No, score=-102.832 tagged_above=-999 required=5 tests=[AWL=-0.233, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id q-CnsGQApYdo for ; Thu, 20 Oct 2011 13:26:22 -0700 (PDT) Received: from darkstar.isi.edu (darkstar.isi.edu [128.9.128.127]) by ietfa.amsl.com (Postfix) with ESMTP id 7EA7811E807F for ; Thu, 20 Oct 2011 13:26:22 -0700 (PDT) Received: from [128.9.160.166] (abc.isi.edu [128.9.160.166]) (authenticated bits=0) by darkstar.isi.edu (8.13.8/8.13.8) with ESMTP id p9KKPmEa005847 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Thu, 20 Oct 2011 13:25:48 -0700 (PDT) Message-ID: <4EA083CC.3070405@isi.edu> Date: Thu, 20 Oct 2011 13:25:48 -0700 From: Joe Touch User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1 MIME-Version: 1.0 To: David Ros References: <4EA05A79.7080700@isi.edu> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: Ingemar Johansson S , "jana.iyengar@gmail.com" , "ledbat@ietf.org" Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 20:26:23 -0000 On 10/20/2011 11:57 AM, David Ros wrote: > > Le 20 oct. 2011 à 19:29, Joe Touch a écrit : > >> >> >> On 10/19/2011 10:21 PM, Ingemar Johansson S wrote: >>> Hi >>> >>> Change of subject name. >>> I have no strong opinion on this but I would like to ask how likely >>> it is that LEDBAT is used for shortlived flows in the first place. >> >> INIT_CWND comes into play during any restart, e.g., after a gap in the offered load. >> > > Isn't that the role of max_allowed_cwnd, instead? That's the highest the window goes to under steady load; INIT is where it starts for new connections and after slow-start restart. Joe From touch@isi.edu Thu Oct 20 13:29:06 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6AC6B11E8090 for ; Thu, 20 Oct 2011 13:29:06 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.799 X-Spam-Level: X-Spam-Status: No, score=-102.799 tagged_above=-999 required=5 tests=[AWL=-0.200, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6rsV0MCHmhcq for ; Thu, 20 Oct 2011 13:29:06 -0700 (PDT) Received: from darkstar.isi.edu (darkstar.isi.edu [128.9.128.127]) by ietfa.amsl.com (Postfix) with ESMTP id F00D311E807F for ; Thu, 20 Oct 2011 13:29:05 -0700 (PDT) Received: from [128.9.160.166] (abc.isi.edu [128.9.160.166]) (authenticated bits=0) by darkstar.isi.edu (8.13.8/8.13.8) with ESMTP id p9KKSJFr007192 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Thu, 20 Oct 2011 13:28:19 -0700 (PDT) Message-ID: <4EA08463.6060208@isi.edu> Date: Thu, 20 Oct 2011 13:28:19 -0700 From: Joe Touch User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1 MIME-Version: 1.0 To: gorry@erg.abdn.ac.uk References: <4EA05A79.7080700@isi.edu> <4EA06532.8040204@erg.abdn.ac.uk> In-Reply-To: <4EA06532.8040204@erg.abdn.ac.uk> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: Ingemar Johansson S , "jana.iyengar@gmail.com" , "ledbat@ietf.org" Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Oct 2011 20:29:06 -0000 On 10/20/2011 11:15 AM, Gorry Fairhurst wrote: > > > On 20/10/2011 18:29, Joe Touch wrote: >> >> On 10/19/2011 10:21 PM, Ingemar Johansson S wrote: >>> Hi >>> >>> Change of subject name. >>> I have no strong opinion on this but I would like to ask how likely >>> it is that LEDBAT is used for shortlived flows in the first place. >> >> INIT_CWND comes into play during any restart, e.g., after a gap in the >> offered load. >> > Is that the "restart window" (RW) in TCP? Yes. Per RFC 2581: For the purposes of this standard, we define RW = IW. LEDBAT doesn't do otherwise, so that's why I've been assuming the same. Joe From ingemar.s.johansson@ericsson.com Thu Oct 20 23:13:21 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D362D11E8095 for ; Thu, 20 Oct 2011 23:13:21 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.599 X-Spam-Level: X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8QpOEtgXlYAa for ; Thu, 20 Oct 2011 23:13:21 -0700 (PDT) Received: from mailgw9.se.ericsson.net (mailgw9.se.ericsson.net [193.180.251.57]) by ietfa.amsl.com (Postfix) with ESMTP id CAE9C11E80B4 for ; Thu, 20 Oct 2011 23:13:20 -0700 (PDT) X-AuditID: c1b4fb39-b7bfdae000005125-fc-4ea10d7d38ba Received: from esessmw0256.eemea.ericsson.se (Unknown_Domain [153.88.253.125]) by mailgw9.se.ericsson.net (Symantec Mail Security) with SMTP id 9D.3B.20773.D7D01AE4; Fri, 21 Oct 2011 08:13:17 +0200 (CEST) Received: from ESESSCMS0366.eemea.ericsson.se ([169.254.1.23]) by esessmw0256.eemea.ericsson.se ([10.2.3.125]) with mapi; Fri, 21 Oct 2011 08:13:17 +0200 From: Ingemar Johansson S To: Joe Touch Date: Fri, 21 Oct 2011 08:13:12 +0200 Thread-Topic: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) Thread-Index: AcyPWhNVRaEix4ILS124Il/EggkAvQAXhWyQ Message-ID: References: <4EA05A79.7080700@isi.edu> In-Reply-To: Accept-Language: sv-SE, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: sv-SE, en-US Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: AAAAAA== Cc: "jana.iyengar@gmail.com" , "ledbat@ietf.org" Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 21 Oct 2011 06:13:21 -0000 OK To me it seems like Initial window =3D 2 is reasonable. Regards Ingemar > -----Original Message----- > From: David Ros [mailto:David.Ros@telecom-bretagne.eu]=20 > Sent: den 20 oktober 2011 20:57 > To: Joe Touch > Cc: Ingemar Johansson S; jana.iyengar@gmail.com; ledbat@ietf.org > Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New=20 > Version Notification for draft-ietf-ledbat-congestion-08.txt) >=20 >=20 > Le 20 oct. 2011 =E0 19:29, Joe Touch a =E9crit : >=20 > >=20 > >=20 > > On 10/19/2011 10:21 PM, Ingemar Johansson S wrote: > >> Hi > >>=20 > >> Change of subject name. > >> I have no strong opinion on this but I would like to ask=20 > how likely=20 > >> it is that LEDBAT is used for shortlived flows in the first place. > >=20 > > INIT_CWND comes into play during any restart, e.g., after a=20 > gap in the offered load. > >=20 >=20 > Isn't that the role of max_allowed_cwnd, instead? >=20 > (section 3.4.2) > This algorithm restricts cwnd growth after a period of inactivity, > where the cwnd is clamped down to a little more than=20 > flightsize using > max_allowed_cwnd >=20 > Thanks, >=20 > David. >=20 > > While I don't expect LEDBAD would be used primarily for=20 > shortlived flows: > > 1) we can't know that, so should plan accordingly > > 2) long-lived flows with frequent gaps do exist, and > > should not cause a problem either > >=20 > > Joe > > _______________________________________________ > > ledbat mailing list > > ledbat@ietf.org > > https://www.ietf.org/mailman/listinfo/ledbat > >=20 >=20 > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > David ROS > http://www.rennes.enst-bretagne.fr/~dros/ >=20 > In theory, theory and practice are the same. In practice,=20 > they are different. -- Author unknown >=20 > = From Rolf.Winter@neclab.eu Tue Oct 25 08:35:36 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CAB8921F8B8D for ; Tue, 25 Oct 2011 08:35:36 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.339 X-Spam-Level: X-Spam-Status: No, score=-102.339 tagged_above=-999 required=5 tests=[AWL=0.260, BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id mJrecaqwRnSj for ; Tue, 25 Oct 2011 08:35:36 -0700 (PDT) Received: from mailer1.neclab.eu (mailer1.neclab.eu [195.37.70.40]) by ietfa.amsl.com (Postfix) with ESMTP id 43D3F21F8B8B for ; Tue, 25 Oct 2011 08:35:31 -0700 (PDT) Received: from localhost (localhost.localdomain [127.0.0.1]) by mailer1.neclab.eu (Postfix) with ESMTP id C5926280001D9 for ; Tue, 25 Oct 2011 17:35:52 +0200 (CEST) X-Virus-Scanned: Amavisd on Debian GNU/Linux (netlab.nec.de) Received: from mailer1.neclab.eu ([127.0.0.1]) by localhost (atlas1.office.hd [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id fQpToJcLJNuO for ; Tue, 25 Oct 2011 17:35:52 +0200 (CEST) Received: from ENCELADUS.office.hd (ENCELADUS.office.hd [192.168.24.52]) by mailer1.neclab.eu (Postfix) with ESMTP id AB4AE28000082 for ; Tue, 25 Oct 2011 17:35:47 +0200 (CEST) Received: from PALLENE.office.hd ([169.254.1.17]) by ENCELADUS.office.hd ([192.168.24.52]) with mapi id 14.01.0323.003; Tue, 25 Oct 2011 17:35:25 +0200 From: Rolf Winter To: "ledbat@ietf.org" Thread-Topic: WG last call on draft-ietf-ledbat-congestion-08 Thread-Index: AcyTK7aJUe55RIMWTOmGwmaJTV07yw== Date: Tue, 25 Oct 2011 15:35:25 +0000 Message-ID: <791AD3077F94194BB2BDD13565B6295D24F4635D@PALLENE.office.hd> Accept-Language: en-US, de-DE Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.7.0.206] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: [ledbat] WG last call on draft-ietf-ledbat-congestion-08 X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 25 Oct 2011 15:35:37 -0000 Hi WG, This last call has now ended. Could the authors comment and/or submit a ver= sion of the document which incorporates the comments made on the list. Thanks a lot, Rolf NEC Europe Limited | Registered Office: NEC House, 1 Victoria Road, London = W3 6BL | Registered in England 2832014=20 From mirja.kuehlewind@ikr.uni-stuttgart.de Fri Oct 28 06:26:47 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1F53521F8A66 for ; Fri, 28 Oct 2011 06:26:47 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.124 X-Spam-Level: X-Spam-Status: No, score=-2.124 tagged_above=-999 required=5 tests=[AWL=0.125, BAYES_00=-2.599, HELO_EQ_DE=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KauJXaLXRt35 for ; Fri, 28 Oct 2011 06:26:46 -0700 (PDT) Received: from mailsrv.ikr.uni-stuttgart.de (mailsrv.ikr.uni-stuttgart.de [129.69.170.2]) by ietfa.amsl.com (Postfix) with ESMTP id 7B71D21F849D for ; Fri, 28 Oct 2011 06:26:46 -0700 (PDT) Received: from netsrv1.ikr.uni-stuttgart.de (netsrv1-c [10.11.12.12]) by mailsrv.ikr.uni-stuttgart.de (Postfix) with ESMTP id 46090633B2; Fri, 28 Oct 2011 15:26:45 +0200 (CEST) Received: from vpn-2-cl177 (vpn-2-cl177 [10.41.21.177]) by netsrv1.ikr.uni-stuttgart.de (Postfix) with ESMTP id 364D459A8A; Fri, 28 Oct 2011 15:26:45 +0200 (CEST) From: Mirja Kuehlewind Organization: University of Stuttgart (Germany), IKR To: ledbat@ietf.org Date: Fri, 28 Oct 2011 15:26:44 +0200 User-Agent: KMail/1.9.10 (enterprise35 0.20101217.1207316) References: <791AD3077F94194BB2BDD13565B6295D1D0A463B@DAPHNIS.office.hd> <32BB54A8-D41C-45C6-949E-F3060A6FF966@telecom-bretagne.eu> In-Reply-To: <32BB54A8-D41C-45C6-949E-F3060A6FF966@telecom-bretagne.eu> X-KMail-QuotePrefix: > MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Message-Id: <201110281526.44554.mirja.kuehlewind@ikr.uni-stuttgart.de> Subject: Re: [ledbat] initialization of current_delays in draft-ietf-ledbat-congestion-08 X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Oct 2011 13:26:47 -0000 Hi David, thanks for your comments. I Incorporated them into the draft.=20 The current_delays initiation thing is an artifact from the previous versio= ns=20 where a filter was used that chooses always the minimum. This was already o= n=20 the ToDo list but we somehow missed it. We want to add a little more text=20 regarding the filter anyway and we might just add a sentence that the=20 infinity values could no be taken into account when applying the filter. Thanks again. Mirja On Thursday 20 October 2011 11:10:51 David Ros wrote: > Hi, > > After reading the current version of the draft, I was wondering (sorry if > it's a very stupid question): > > Why does the current_delays list have to be initialized to +INFINITY? ("on > initialization:", sec. 3.4.2) I understand why that's the case for > base_delays, but not for current_delays. > > Thanks, > > David. > > PS: Jana and Mirja, I'll send you offline a list of typos I found. > > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > David ROS > http://www.rennes.enst-bretagne.fr/~dros/ > > "But we're a university! [...] What sort of people would we be if we didn= 't > go into the Library?" "Students." -- Terry Pratchett > > > > > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat =2D-=20 =2D------------------------------------------------------------------ Dipl.-Ing. Mirja K=FChlewind Institute of Communication Networks and Computer Engineering (IKR) University of Stuttgart, Germany Pfaffenwaldring 47, D-70569 Stuttgart tel: +49(0)711/685-67973 email: mirja.kuehlewind@ikr.uni-stuttgart.de web: www.ikr.uni-stuttgart.de =2D------------------------------------------------------------------ From mirja.kuehlewind@ikr.uni-stuttgart.de Fri Oct 28 06:28:55 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DAF1821F8B19 for ; Fri, 28 Oct 2011 06:28:55 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.142 X-Spam-Level: X-Spam-Status: No, score=-2.142 tagged_above=-999 required=5 tests=[AWL=0.107, BAYES_00=-2.599, HELO_EQ_DE=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id M32PrtWIFNCd for ; Fri, 28 Oct 2011 06:28:55 -0700 (PDT) Received: from mailsrv.ikr.uni-stuttgart.de (mailsrv.ikr.uni-stuttgart.de [129.69.170.2]) by ietfa.amsl.com (Postfix) with ESMTP id 1D84021F8A80 for ; Fri, 28 Oct 2011 06:28:54 -0700 (PDT) Received: from netsrv1.ikr.uni-stuttgart.de (netsrv1-c [10.11.12.12]) by mailsrv.ikr.uni-stuttgart.de (Postfix) with ESMTP id 777BB633B2; Fri, 28 Oct 2011 15:28:53 +0200 (CEST) Received: from vpn-2-cl177 (vpn-2-cl177 [10.41.21.177]) by netsrv1.ikr.uni-stuttgart.de (Postfix) with ESMTP id 6AC2159A8A; Fri, 28 Oct 2011 15:28:53 +0200 (CEST) From: Mirja Kuehlewind Organization: University of Stuttgart (Germany), IKR To: ledbat@ietf.org Date: Fri, 28 Oct 2011 15:28:52 +0200 User-Agent: KMail/1.9.10 (enterprise35 0.20101217.1207316) References: In-Reply-To: X-KMail-QuotePrefix: > MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Message-Id: <201110281528.52801.mirja.kuehlewind@ikr.uni-stuttgart.de> Cc: Ingemar Johansson S , "jana.iyengar@gmail.com" Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New Version Notification for draft-ietf-ledbat-congestion-08.txt) X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Oct 2011 13:28:56 -0000 Hi all, thanks for you feedback. We will change the INIT_CWND in the draft.=20 =46urthermore we have to check if this has any consequences for the part=20 talking about the filter. But I guess this issue can be regarded as solved= =20 now. Mirja On Friday 21 October 2011 08:13:12 Ingemar Johansson S wrote: > OK > > To me it seems like Initial window =3D 2 is reasonable. > > Regards > Ingemar > > > -----Original Message----- > > From: David Ros [mailto:David.Ros@telecom-bretagne.eu] > > Sent: den 20 oktober 2011 20:57 > > To: Joe Touch > > Cc: Ingemar Johansson S; jana.iyengar@gmail.com; ledbat@ietf.org > > Subject: Re: [ledbat] INIT_CWND and MIN_CWND (WAS Re: New > > Version Notification for draft-ietf-ledbat-congestion-08.txt) > > > > Le 20 oct. 2011 =E0 19:29, Joe Touch a =E9crit : > > > On 10/19/2011 10:21 PM, Ingemar Johansson S wrote: > > >> Hi > > >> > > >> Change of subject name. > > >> I have no strong opinion on this but I would like to ask > > > > how likely > > > > >> it is that LEDBAT is used for shortlived flows in the first place. > > > > > > INIT_CWND comes into play during any restart, e.g., after a > > > > gap in the offered load. > > > > > > Isn't that the role of max_allowed_cwnd, instead? > > > > (section 3.4.2) > > This algorithm restricts cwnd growth after a period of inactivity, > > where the cwnd is clamped down to a little more than > > flightsize using > > max_allowed_cwnd > > > > Thanks, > > > > David. > > > > > While I don't expect LEDBAD would be used primarily for > > > > shortlived flows: > > > 1) we can't know that, so should plan accordingly > > > 2) long-lived flows with frequent gaps do exist, and > > > should not cause a problem either > > > > > > Joe > > > _______________________________________________ > > > ledbat mailing list > > > ledbat@ietf.org > > > https://www.ietf.org/mailman/listinfo/ledbat > > > > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > > David ROS > > http://www.rennes.enst-bretagne.fr/~dros/ > > > > In theory, theory and practice are the same. In practice, > > they are different. -- Author unknown > > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat =2D-=20 =2D------------------------------------------------------------------ Dipl.-Ing. Mirja K=FChlewind Institute of Communication Networks and Computer Engineering (IKR) University of Stuttgart, Germany Pfaffenwaldring 47, D-70569 Stuttgart tel: +49(0)711/685-67973 email: mirja.kuehlewind@ikr.uni-stuttgart.de web: www.ikr.uni-stuttgart.de =2D------------------------------------------------------------------ From mirja.kuehlewind@ikr.uni-stuttgart.de Fri Oct 28 06:37:40 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id DCD9D21F8B84 for ; Fri, 28 Oct 2011 06:37:40 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.155 X-Spam-Level: X-Spam-Status: No, score=-2.155 tagged_above=-999 required=5 tests=[AWL=0.094, BAYES_00=-2.599, HELO_EQ_DE=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pligSyPlLoBR for ; Fri, 28 Oct 2011 06:37:35 -0700 (PDT) Received: from mailsrv.ikr.uni-stuttgart.de (mailsrv.ikr.uni-stuttgart.de [129.69.170.2]) by ietfa.amsl.com (Postfix) with ESMTP id AC7DC21F8B64 for ; Fri, 28 Oct 2011 06:37:34 -0700 (PDT) Received: from netsrv1.ikr.uni-stuttgart.de (netsrv1-c [10.11.12.12]) by mailsrv.ikr.uni-stuttgart.de (Postfix) with ESMTP id CDBAB633B2; Fri, 28 Oct 2011 15:37:33 +0200 (CEST) Received: from vpn-2-cl177 (vpn-2-cl177 [10.41.21.177]) by netsrv1.ikr.uni-stuttgart.de (Postfix) with ESMTP id BD2C659A8A; Fri, 28 Oct 2011 15:37:33 +0200 (CEST) From: Mirja Kuehlewind Organization: University of Stuttgart (Germany), IKR To: ledbat@ietf.org Date: Fri, 28 Oct 2011 15:37:33 +0200 User-Agent: KMail/1.9.10 (enterprise35 0.20101217.1207316) References: <4E92338F.1030601@fandm.edu> In-Reply-To: X-KMail-QuotePrefix: > MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Message-Id: <201110281537.33106.mirja.kuehlewind@ikr.uni-stuttgart.de> Cc: adithya kumar , "Dr. R.Leela Velusamy" Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Oct 2011 13:37:41 -0000 Hi Arjuna, heavy congestion does mean that for CTO time no data have been received. If= =20 data was received LEDBAT will compare the respective delay measurement with= =20 the base delay and increase or decrease the cwnd (except it is already=20 MIN_CWND). Then if less packets are in flight than the cwnd allows it will= =20 send out a packet. Normally, if in CTO time no packet was received, it is=20 assumed that all packets are loss and fightsize might be reset to 0. But an= =20 algorithm how to detect packet loss is independent from LEDBAT and thus not= =20 described in this draft. I hope that helps...? Thanks for your feedback! Mirja On Saturday 15 October 2011 10:37:06 Arjuna Sathiaseelan wrote: > Dear Jana and others, > > =A0I went through the draft and am currently trying to figure out what > the performance implications are with the latest draft through > simulations (I have a few helpful UG students in India who have > volunteered to do this for me - this includes modifying the current > ns-2 code to reflect the latest draft). But unfortunately, I dont > think we will be able to tell anything useful before WGLC ends! > > However I have one question -> > =A0" A CTO is used to detect heavy congestion indicated by loss of all > =A0 outstanding data or acknowledgments, resulting in reduction of the > =A0 cwnd to 1 MSS and an exponential backoff of the CTO interval" > > So I am trying to see =A0whether heavy congestion actually means loss of > "ALL" outstanding segments - so how does LEDBAT behave according to > the current draft if atleast one packet went through? Just trying to > understand. > > Regards > Arjuna > > On 10 October 2011 00:51, Janardhan Iyengar wrot= e: > > Dear all, > > > > A new version of the congestion control draft is in the repository. The= re > > are two major mods in this revision: > > > > 1/ WGLC identified one major issue that needed to be addressed in the > > LEDBAT congestion control draft -- LEDBAT response to extreme congestion > > -- and we've tried to address that issue in this revision. We've added a > > new mechanism, the Congestion Timeout (CTO), for a sender to respond to > > extreme congestion. We do not specify how this should be implemented, b= ut > > we do note that a CTO can be implemented with or without a timer. In > > terms of textual changes, we've added an update_CTO() function and a > > branch for what to do if no acks are received within a CTO amount of ti= me > > in Section 3.4.2. We have changed the response to data loss to ensure > > that a protocol, such as TCP, that uses the same timer for both > > congestion control and for > > retransmissions, changes its cwnd correctly. > > > > 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and have > > clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in > > Section 3.5. > > > > Please comment! > > - jana > > > > -- > > Janardhan Iyengar > > Assistant Professor, Computer Science > > Franklin & Marshall College > > http://www.fandm.edu/jiyengar > > > > > > -------- Original Message -------- > > Subject: New Version Notification for draft-ietf-ledbat-congestion-08.t= xt > > Date: Sun, 09 Oct 2011 16:33:33 -0700 > > From: internet-drafts@ietf.org > > To: jiyengar@fandm.edu > > CC: jiyengar@fandm.edu, mirja.kuehlewind@ikr.uni-stuttgart.de, > > =A0greg@bittorrent.com, shalunov@bittorrent.com > > > > A new version of I-D, draft-ietf-ledbat-congestion-08.txt has been > > successfully submitted by Janardhan Iyengar and posted to the IETF > > repository. > > > > Filename: =A0 =A0 =A0 =A0draft-ietf-ledbat-congestion > > Revision: =A0 =A0 =A0 =A008 > > Title: =A0 =A0 =A0 =A0 =A0 Low Extra Delay Background Transport (LEDBAT) > > Creation date: =A0 2011-10-09 > > WG ID: =A0 =A0 =A0 =A0 =A0 ledbat > > Number of pages: 19 > > > > Abstract: > > =A0 LEDBAT is an experimental delay-based congestion control algorithm > > =A0 that attempts to utilize the available bandwidth on an end-to-end > > =A0 path while limiting the consequent increase in queueing delay on the > > =A0 path. =A0LEDBAT uses changes in one-way delay measurements to limit > > =A0 congestion that the flow itself induces in the network. =A0LEDBAT is > > =A0 designed for use by background bulk-transfer applications; it is > > =A0 designed to be no more aggressive than TCP congestion control and to > > =A0 yield in the presence of any competing flows when latency builds, > > =A0 thus limiting interference with the network performance of the > > =A0 competing flows. > > > > > > > > > > The IETF Secretariat > > _______________________________________________ > > ledbat mailing list > > ledbat@ietf.org > > https://www.ietf.org/mailman/listinfo/ledbat > > -- > http://about.me/arjuna.sathiaseelan > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat =2D-=20 =2D------------------------------------------------------------------ Dipl.-Ing. Mirja K=FChlewind Institute of Communication Networks and Computer Engineering (IKR) University of Stuttgart, Germany Pfaffenwaldring 47, D-70569 Stuttgart tel: +49(0)711/685-67973 email: mirja.kuehlewind@ikr.uni-stuttgart.de web: www.ikr.uni-stuttgart.de =2D------------------------------------------------------------------ From mirja.kuehlewind@ikr.uni-stuttgart.de Fri Oct 28 06:52:53 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B6A0221F8634 for ; Fri, 28 Oct 2011 06:52:53 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.166 X-Spam-Level: X-Spam-Status: No, score=-2.166 tagged_above=-999 required=5 tests=[AWL=0.083, BAYES_00=-2.599, HELO_EQ_DE=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id z-EnISOQBCuL for ; Fri, 28 Oct 2011 06:52:53 -0700 (PDT) Received: from mailsrv.ikr.uni-stuttgart.de (mailsrv.ikr.uni-stuttgart.de [129.69.170.2]) by ietfa.amsl.com (Postfix) with ESMTP id 9D82021F862F for ; Fri, 28 Oct 2011 06:52:52 -0700 (PDT) Received: from netsrv1.ikr.uni-stuttgart.de (netsrv1-c [10.11.12.12]) by mailsrv.ikr.uni-stuttgart.de (Postfix) with ESMTP id C17CB633B2; Fri, 28 Oct 2011 15:52:51 +0200 (CEST) Received: from vpn-2-cl177 (vpn-2-cl177 [10.41.21.177]) by netsrv1.ikr.uni-stuttgart.de (Postfix) with ESMTP id B26C059A8A; Fri, 28 Oct 2011 15:52:51 +0200 (CEST) From: Mirja Kuehlewind Organization: University of Stuttgart (Germany), IKR To: ledbat@ietf.org Date: Fri, 28 Oct 2011 15:52:50 +0200 User-Agent: KMail/1.9.10 (enterprise35 0.20101217.1207316) References: <4E92338F.1030601@fandm.edu> In-Reply-To: X-KMail-QuotePrefix: > MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Message-Id: <201110281552.50770.mirja.kuehlewind@ikr.uni-stuttgart.de> Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Oct 2011 13:52:53 -0000 Thanks, Arjuna. I added two sentences here: "LEDBAT takes the minimum delay it measures at any time as the base delay a= nd=20 compares the current delay measurement to this value to calculate the queui= ng=20 delay. off_target compares the queuing delay to the dedicated TARGET chosen= =20 and is normalizes by the TARGET value itself." Does this work? Mirja On Saturday 15 October 2011 10:56:49 Arjuna Sathiaseelan wrote: > Another thing which I forgot to point out - in section 3.4.1 the term > off_target suddenly pops up without an explanation. That needs to be > resolved please. > > Regards > Arjuna > > On 15 October 2011 09:37, Arjuna Sathiaseelan wro= te: > > Dear Jana and others, > > > > =A0I went through the draft and am currently trying to figure out what > > the performance implications are with the latest draft through > > simulations (I have a few helpful UG students in India who have > > volunteered to do this for me - this includes modifying the current > > ns-2 code to reflect the latest draft). But unfortunately, I dont > > think we will be able to tell anything useful before WGLC ends! > > > > However I have one question -> > > =A0" A CTO is used to detect heavy congestion indicated by loss of all > > =A0 outstanding data or acknowledgments, resulting in reduction of the > > =A0 cwnd to 1 MSS and an exponential backoff of the CTO interval" > > > > So I am trying to see =A0whether heavy congestion actually means loss of > > "ALL" outstanding segments - so how does LEDBAT behave according to > > the current draft if atleast one packet went through? Just trying to > > understand. > > > > Regards > > Arjuna > > > > On 10 October 2011 00:51, Janardhan Iyengar =20 wrote: > >> Dear all, > >> > >> A new version of the congestion control draft is in the repository. > >> There are two major mods in this revision: > >> > >> 1/ WGLC identified one major issue that needed to be addressed in the > >> LEDBAT congestion control draft -- LEDBAT response to extreme congesti= on > >> -- and we've tried to address that issue in this revision. We've added= a > >> new mechanism, the Congestion Timeout (CTO), for a sender to respond to > >> extreme congestion. We do not specify how this should be implemented, > >> but we do note that a CTO can be implemented with or without a timer. = In > >> terms of textual changes, we've added an update_CTO() function and a > >> branch for what to do if no acks are received within a CTO amount of > >> time in Section 3.4.2. We have changed the response to data loss to > >> ensure that a protocol, such as TCP, that uses the same timer for both > >> congestion control and for > >> retransmissions, changes its cwnd correctly. > >> > >> 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and ha= ve > >> clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in > >> Section 3.5. > >> > >> Please comment! > >> - jana > >> > >> -- > >> Janardhan Iyengar > >> Assistant Professor, Computer Science > >> Franklin & Marshall College > >> http://www.fandm.edu/jiyengar > >> > >> > >> -------- Original Message -------- > >> Subject: New Version Notification for > >> draft-ietf-ledbat-congestion-08.txt Date: Sun, 09 Oct 2011 16:33:33 > >> -0700 > >> From: internet-drafts@ietf.org > >> To: jiyengar@fandm.edu > >> CC: jiyengar@fandm.edu, mirja.kuehlewind@ikr.uni-stuttgart.de, > >> =A0greg@bittorrent.com, shalunov@bittorrent.com > >> > >> A new version of I-D, draft-ietf-ledbat-congestion-08.txt has been > >> successfully submitted by Janardhan Iyengar and posted to the IETF > >> repository. > >> > >> Filename: =A0 =A0 =A0 =A0draft-ietf-ledbat-congestion > >> Revision: =A0 =A0 =A0 =A008 > >> Title: =A0 =A0 =A0 =A0 =A0 Low Extra Delay Background Transport (LEDBA= T) > >> Creation date: =A0 2011-10-09 > >> WG ID: =A0 =A0 =A0 =A0 =A0 ledbat > >> Number of pages: 19 > >> > >> Abstract: > >> =A0 LEDBAT is an experimental delay-based congestion control algorithm > >> =A0 that attempts to utilize the available bandwidth on an end-to-end > >> =A0 path while limiting the consequent increase in queueing delay on t= he > >> =A0 path. =A0LEDBAT uses changes in one-way delay measurements to limit > >> =A0 congestion that the flow itself induces in the network. =A0LEDBAT = is > >> =A0 designed for use by background bulk-transfer applications; it is > >> =A0 designed to be no more aggressive than TCP congestion control and = to > >> =A0 yield in the presence of any competing flows when latency builds, > >> =A0 thus limiting interference with the network performance of the > >> =A0 competing flows. > >> > >> > >> > >> > >> The IETF Secretariat > >> _______________________________________________ > >> ledbat mailing list > >> ledbat@ietf.org > >> https://www.ietf.org/mailman/listinfo/ledbat > > > > -- > > http://about.me/arjuna.sathiaseelan > > _______________________________________________ > ledbat mailing list > ledbat@ietf.org > https://www.ietf.org/mailman/listinfo/ledbat =2D-=20 =2D------------------------------------------------------------------ Dipl.-Ing. Mirja K=FChlewind Institute of Communication Networks and Computer Engineering (IKR) University of Stuttgart, Germany Pfaffenwaldring 47, D-70569 Stuttgart tel: +49(0)711/685-67973 email: mirja.kuehlewind@ikr.uni-stuttgart.de web: www.ikr.uni-stuttgart.de =2D------------------------------------------------------------------ From arjuna.sathiaseelan@gmail.com Fri Oct 28 09:10:55 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1CD1321F89BA for ; Fri, 28 Oct 2011 09:10:55 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.977 X-Spam-Level: X-Spam-Status: No, score=-2.977 tagged_above=-999 required=5 tests=[AWL=-0.000, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id hRkCOj21Euoe for ; Fri, 28 Oct 2011 09:10:48 -0700 (PDT) Received: from mail-qw0-f44.google.com (mail-qw0-f44.google.com [209.85.216.44]) by ietfa.amsl.com (Postfix) with ESMTP id B7DEC21F8B00 for ; Fri, 28 Oct 2011 09:10:47 -0700 (PDT) Received: by qadc10 with SMTP id c10so4785929qad.31 for ; Fri, 28 Oct 2011 09:10:31 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=nDMNwPo6PQNNlszQYZ1SiUQSHI1GF+7e92Fw9XLTivk=; b=DOYDNAgk+Ga+uGzdRHLZdy+5Ow/AO7PF0k2CmQujxKCF2aG3xtlR31/EseSnzIq6YI GwyvNy4ydLXjsNq/X1naUVKNs2CbRRnWUPe/6cOvPMVNxRXwEI9FDUGyDsDgWF+HAnoz P7nUjZCYJ7sMv2IX69bQbx5dGWcqYxYC+F6q0= MIME-Version: 1.0 Received: by 10.229.13.11 with SMTP id z11mr870134qcz.115.1319818230965; Fri, 28 Oct 2011 09:10:30 -0700 (PDT) Sender: arjuna.sathiaseelan@gmail.com Received: by 10.229.28.196 with HTTP; Fri, 28 Oct 2011 09:10:30 -0700 (PDT) In-Reply-To: <201110281552.50770.mirja.kuehlewind@ikr.uni-stuttgart.de> References: <4E92338F.1030601@fandm.edu> <201110281552.50770.mirja.kuehlewind@ikr.uni-stuttgart.de> Date: Fri, 28 Oct 2011 17:10:30 +0100 X-Google-Sender-Auth: XSkoh2dC7ZyZu2Ld8bTZyZaT_k0 Message-ID: From: Arjuna Sathiaseelan To: Mirja Kuehlewind Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: ledbat@ietf.org Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Oct 2011 16:10:55 -0000 Hello Mirja, How about this? "LEDBAT takes the minimum delay it measures at any time as the base delay a= nd compares the current delay measurement to this value to calculate the queui= ng delay. off_target compares the queuing delay to the chosen TARGET delay and normalizes itself to the TARGET value." Regards Arjuna On 28 October 2011 14:52, Mirja Kuehlewind wrote: > Thanks, Arjuna. > > I added two sentences here: > "LEDBAT takes the minimum delay it measures at any time as the base delay= and > compares the current delay measurement to this value to calculate the que= uing > delay. off_target compares the queuing delay to the dedicated TARGET chos= en > and is normalizes by the TARGET value itself." > > Does this work? > > Mirja > > > On Saturday 15 October 2011 10:56:49 Arjuna Sathiaseelan wrote: >> Another thing which I forgot to point out - in section 3.4.1 the term >> off_target suddenly pops up without an explanation. That needs to be >> resolved please. >> >> Regards >> Arjuna >> >> On 15 October 2011 09:37, Arjuna Sathiaseelan wr= ote: >> > Dear Jana and others, >> > >> > =A0I went through the draft and am currently trying to figure out what >> > the performance implications are with the latest draft through >> > simulations (I have a few helpful UG students in India who have >> > volunteered to do this for me - this includes modifying the current >> > ns-2 code to reflect the latest draft). But unfortunately, I dont >> > think we will be able to tell anything useful before WGLC ends! >> > >> > However I have one question -> >> > =A0" A CTO is used to detect heavy congestion indicated by loss of all >> > =A0 outstanding data or acknowledgments, resulting in reduction of the >> > =A0 cwnd to 1 MSS and an exponential backoff of the CTO interval" >> > >> > So I am trying to see =A0whether heavy congestion actually means loss = of >> > "ALL" outstanding segments - so how does LEDBAT behave according to >> > the current draft if atleast one packet went through? Just trying to >> > understand. >> > >> > Regards >> > Arjuna >> > >> > On 10 October 2011 00:51, Janardhan Iyengar > wrote: >> >> Dear all, >> >> >> >> A new version of the congestion control draft is in the repository. >> >> There are two major mods in this revision: >> >> >> >> 1/ WGLC identified one major issue that needed to be addressed in the >> >> LEDBAT congestion control draft -- LEDBAT response to extreme congest= ion >> >> -- and we've tried to address that issue in this revision. We've adde= d a >> >> new mechanism, the Congestion Timeout (CTO), for a sender to respond = to >> >> extreme congestion. We do not specify how this should be implemented, >> >> but we do note that a CTO can be implemented with or without a timer.= In >> >> terms of textual changes, we've added an update_CTO() function and a >> >> branch for what to do if no acks are received within a CTO amount of >> >> time in Section 3.4.2. We have changed the response to data loss to >> >> ensure that a protocol, such as TCP, that uses the same timer for bot= h >> >> congestion control and for >> >> retransmissions, changes its cwnd correctly. >> >> >> >> 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and h= ave >> >> clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in >> >> Section 3.5. >> >> >> >> Please comment! >> >> - jana >> >> >> >> -- >> >> Janardhan Iyengar >> >> Assistant Professor, Computer Science >> >> Franklin & Marshall College >> >> http://www.fandm.edu/jiyengar >> >> >> >> >> >> -------- Original Message -------- >> >> Subject: New Version Notification for >> >> draft-ietf-ledbat-congestion-08.txt Date: Sun, 09 Oct 2011 16:33:33 >> >> -0700 >> >> From: internet-drafts@ietf.org >> >> To: jiyengar@fandm.edu >> >> CC: jiyengar@fandm.edu, mirja.kuehlewind@ikr.uni-stuttgart.de, >> >> =A0greg@bittorrent.com, shalunov@bittorrent.com >> >> >> >> A new version of I-D, draft-ietf-ledbat-congestion-08.txt has been >> >> successfully submitted by Janardhan Iyengar and posted to the IETF >> >> repository. >> >> >> >> Filename: =A0 =A0 =A0 =A0draft-ietf-ledbat-congestion >> >> Revision: =A0 =A0 =A0 =A008 >> >> Title: =A0 =A0 =A0 =A0 =A0 Low Extra Delay Background Transport (LEDB= AT) >> >> Creation date: =A0 2011-10-09 >> >> WG ID: =A0 =A0 =A0 =A0 =A0 ledbat >> >> Number of pages: 19 >> >> >> >> Abstract: >> >> =A0 LEDBAT is an experimental delay-based congestion control algorith= m >> >> =A0 that attempts to utilize the available bandwidth on an end-to-end >> >> =A0 path while limiting the consequent increase in queueing delay on = the >> >> =A0 path. =A0LEDBAT uses changes in one-way delay measurements to lim= it >> >> =A0 congestion that the flow itself induces in the network. =A0LEDBAT= is >> >> =A0 designed for use by background bulk-transfer applications; it is >> >> =A0 designed to be no more aggressive than TCP congestion control and= to >> >> =A0 yield in the presence of any competing flows when latency builds, >> >> =A0 thus limiting interference with the network performance of the >> >> =A0 competing flows. >> >> >> >> >> >> >> >> >> >> The IETF Secretariat >> >> _______________________________________________ >> >> ledbat mailing list >> >> ledbat@ietf.org >> >> https://www.ietf.org/mailman/listinfo/ledbat >> > >> > -- >> > http://about.me/arjuna.sathiaseelan >> >> _______________________________________________ >> ledbat mailing list >> ledbat@ietf.org >> https://www.ietf.org/mailman/listinfo/ledbat > > > > -- > ------------------------------------------------------------------- > Dipl.-Ing. Mirja K=FChlewind > Institute of Communication Networks and Computer Engineering (IKR) > University of Stuttgart, Germany > Pfaffenwaldring 47, D-70569 Stuttgart > > tel: +49(0)711/685-67973 > email: mirja.kuehlewind@ikr.uni-stuttgart.de > web: www.ikr.uni-stuttgart.de > ------------------------------------------------------------------- > From arjuna.sathiaseelan@gmail.com Fri Oct 28 09:12:44 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4981321F8B15 for ; Fri, 28 Oct 2011 09:12:44 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.977 X-Spam-Level: X-Spam-Status: No, score=-2.977 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id dNqcV7hcEoJG for ; Fri, 28 Oct 2011 09:12:43 -0700 (PDT) Received: from mail-qw0-f44.google.com (mail-qw0-f44.google.com [209.85.216.44]) by ietfa.amsl.com (Postfix) with ESMTP id 4BEE821F8AFE for ; Fri, 28 Oct 2011 09:12:43 -0700 (PDT) Received: by qadc10 with SMTP id c10so4788624qad.31 for ; Fri, 28 Oct 2011 09:12:42 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:sender:in-reply-to:references:date :x-google-sender-auth:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=gAr/4SXMYIwpc/u1D6ja2v34JeWTQ5bXNZXOKYSQGus=; b=SeS4lUL3BbyoFNUtPLFGWXc/tHMusjHRriNHDa0Pb2T4g2ph0gpAkpg3P5u6Van9/R GQLk0RbkVt6AtuawIfiOh+kXKW5HGuZAbNBpm7wk/en0xQRQvOs58bdKbAFbOLS45mAS OkUS3RV6vpKhqATpL+52nCKbIXNUkhyyQvEQg= MIME-Version: 1.0 Received: by 10.229.74.79 with SMTP id t15mr863040qcj.153.1319818362475; Fri, 28 Oct 2011 09:12:42 -0700 (PDT) Sender: arjuna.sathiaseelan@gmail.com Received: by 10.229.28.196 with HTTP; Fri, 28 Oct 2011 09:12:42 -0700 (PDT) In-Reply-To: <201110281537.33106.mirja.kuehlewind@ikr.uni-stuttgart.de> References: <4E92338F.1030601@fandm.edu> <201110281537.33106.mirja.kuehlewind@ikr.uni-stuttgart.de> Date: Fri, 28 Oct 2011 17:12:42 +0100 X-Google-Sender-Auth: 01NZvUuavrpZ4Jmd6rlOJLXfGCg Message-ID: From: Arjuna Sathiaseelan To: Mirja Kuehlewind Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: adithya kumar , ledbat@ietf.org, "Dr. R.Leela Velusamy" Subject: Re: [ledbat] Fwd: New Version Notification for draft-ietf-ledbat-congestion-08.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 28 Oct 2011 16:12:44 -0000 Thanks Mirja - that helps :).. Regards Arjuna On 28 October 2011 14:37, Mirja Kuehlewind wrote: > Hi Arjuna, > > heavy congestion does mean that for CTO time no data have been received. = If > data was received LEDBAT will compare the respective delay measurement wi= th > the base delay and increase or decrease the cwnd (except it is already > MIN_CWND). Then if less packets are in flight than the cwnd allows it wil= l > send out a packet. Normally, if in CTO time no packet was received, it is > assumed that all packets are loss and fightsize might be reset to 0. But = an > algorithm how to detect packet loss is independent from LEDBAT and thus n= ot > described in this draft. > > I hope that helps...? > > Thanks for your feedback! > Mirja > > > On Saturday 15 October 2011 10:37:06 Arjuna Sathiaseelan wrote: >> Dear Jana and others, >> >> =A0I went through the draft and am currently trying to figure out what >> the performance implications are with the latest draft through >> simulations (I have a few helpful UG students in India who have >> volunteered to do this for me - this includes modifying the current >> ns-2 code to reflect the latest draft). But unfortunately, I dont >> think we will be able to tell anything useful before WGLC ends! >> >> However I have one question -> >> =A0" A CTO is used to detect heavy congestion indicated by loss of all >> =A0 outstanding data or acknowledgments, resulting in reduction of the >> =A0 cwnd to 1 MSS and an exponential backoff of the CTO interval" >> >> So I am trying to see =A0whether heavy congestion actually means loss of >> "ALL" outstanding segments - so how does LEDBAT behave according to >> the current draft if atleast one packet went through? Just trying to >> understand. >> >> Regards >> Arjuna >> >> On 10 October 2011 00:51, Janardhan Iyengar wro= te: >> > Dear all, >> > >> > A new version of the congestion control draft is in the repository. Th= ere >> > are two major mods in this revision: >> > >> > 1/ WGLC identified one major issue that needed to be addressed in the >> > LEDBAT congestion control draft -- LEDBAT response to extreme congesti= on >> > -- and we've tried to address that issue in this revision. We've added= a >> > new mechanism, the Congestion Timeout (CTO), for a sender to respond t= o >> > extreme congestion. We do not specify how this should be implemented, = but >> > we do note that a CTO can be implemented with or without a timer. In >> > terms of textual changes, we've added an update_CTO() function and a >> > branch for what to do if no acks are received within a CTO amount of t= ime >> > in Section 3.4.2. We have changed the response to data loss to ensure >> > that a protocol, such as TCP, that uses the same timer for both >> > congestion control and for >> > retransmissions, changes its cwnd correctly. >> > >> > 2/ We have set the values for INIT_CWND to 4 and MIN_CWND to 2, and ha= ve >> > clarified the discussion of CURRENT_DELAYS and INIT_CWND/MIN_CWND in >> > Section 3.5. >> > >> > Please comment! >> > - jana >> > >> > -- >> > Janardhan Iyengar >> > Assistant Professor, Computer Science >> > Franklin & Marshall College >> > http://www.fandm.edu/jiyengar >> > >> > >> > -------- Original Message -------- >> > Subject: New Version Notification for draft-ietf-ledbat-congestion-08.= txt >> > Date: Sun, 09 Oct 2011 16:33:33 -0700 >> > From: internet-drafts@ietf.org >> > To: jiyengar@fandm.edu >> > CC: jiyengar@fandm.edu, mirja.kuehlewind@ikr.uni-stuttgart.de, >> > =A0greg@bittorrent.com, shalunov@bittorrent.com >> > >> > A new version of I-D, draft-ietf-ledbat-congestion-08.txt has been >> > successfully submitted by Janardhan Iyengar and posted to the IETF >> > repository. >> > >> > Filename: =A0 =A0 =A0 =A0draft-ietf-ledbat-congestion >> > Revision: =A0 =A0 =A0 =A008 >> > Title: =A0 =A0 =A0 =A0 =A0 Low Extra Delay Background Transport (LEDBA= T) >> > Creation date: =A0 2011-10-09 >> > WG ID: =A0 =A0 =A0 =A0 =A0 ledbat >> > Number of pages: 19 >> > >> > Abstract: >> > =A0 LEDBAT is an experimental delay-based congestion control algorithm >> > =A0 that attempts to utilize the available bandwidth on an end-to-end >> > =A0 path while limiting the consequent increase in queueing delay on t= he >> > =A0 path. =A0LEDBAT uses changes in one-way delay measurements to limi= t >> > =A0 congestion that the flow itself induces in the network. =A0LEDBAT = is >> > =A0 designed for use by background bulk-transfer applications; it is >> > =A0 designed to be no more aggressive than TCP congestion control and = to >> > =A0 yield in the presence of any competing flows when latency builds, >> > =A0 thus limiting interference with the network performance of the >> > =A0 competing flows. >> > >> > >> > >> > >> > The IETF Secretariat >> > _______________________________________________ >> > ledbat mailing list >> > ledbat@ietf.org >> > https://www.ietf.org/mailman/listinfo/ledbat >> >> -- >> http://about.me/arjuna.sathiaseelan >> _______________________________________________ >> ledbat mailing list >> ledbat@ietf.org >> https://www.ietf.org/mailman/listinfo/ledbat > > > > -- > ------------------------------------------------------------------- > Dipl.-Ing. Mirja K=FChlewind > Institute of Communication Networks and Computer Engineering (IKR) > University of Stuttgart, Germany > Pfaffenwaldring 47, D-70569 Stuttgart > > tel: +49(0)711/685-67973 > email: mirja.kuehlewind@ikr.uni-stuttgart.de > web: www.ikr.uni-stuttgart.de > ------------------------------------------------------------------- > From jana.iyengar@gmail.com Mon Oct 31 12:54:43 2011 Return-Path: X-Original-To: ledbat@ietfa.amsl.com Delivered-To: ledbat@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A149C11E81C6 for ; Mon, 31 Oct 2011 12:54:43 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.599 X-Spam-Level: X-Spam-Status: No, score=-3.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id H7EiSm1hlcwt for ; Mon, 31 Oct 2011 12:54:43 -0700 (PDT) Received: from mail-pz0-f50.google.com (mail-pz0-f50.google.com [209.85.210.50]) by ietfa.amsl.com (Postfix) with ESMTP id 5B76111E81A8 for ; Mon, 31 Oct 2011 12:54:40 -0700 (PDT) Received: by pzk34 with SMTP id 34so18247043pzk.9 for ; Mon, 31 Oct 2011 12:54:40 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=message-id:date:from:reply-to:organization:user-agent:mime-version :to:cc:subject:references:in-reply-to:x-forwarded-message-id :content-type:content-transfer-encoding; bh=NfEMtz5qkOqENkQeW3OefrttMN6dluFJsk4l3PGfhV4=; b=p/RlaEdhvegiLLWD4i8Qv2EHEfQY75Lu3QNNzS/E9SCcTBW0bx7Nt50M1VD+Oma73E +NvDfjo5RxXzTQeMiRpQ98RnmvyI317jRV/zxi+3mLL90X2wWdxlIJLh2jq1KXA+KkKD ZgB8I38YxOH2IDeYPnIHw4sfPR9KRpstsSgqg= Received: by 10.68.34.2 with SMTP id v2mr9764513pbi.112.1320090880030; Mon, 31 Oct 2011 12:54:40 -0700 (PDT) Received: from surutti.fandm.edu ([2620:0:1b00:131f:b9e8:e56f:d1b8:5db1]) by mx.google.com with ESMTPS id 3sm52811707pbb.19.2011.10.31.12.54.37 (version=TLSv1/SSLv3 cipher=OTHER); Mon, 31 Oct 2011 12:54:38 -0700 (PDT) Message-ID: <4EAEFCFC.9060600@fandm.edu> Date: Mon, 31 Oct 2011 12:54:36 -0700 From: Janardhan Iyengar Organization: Franklin & Marshall College User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:7.0.1) Gecko/20110929 Thunderbird/7.0.1 MIME-Version: 1.0 To: ledbat@ietf.org References: <20111031194200.18624.29794.idtracker@ietfa.amsl.com> In-Reply-To: <20111031194200.18624.29794.idtracker@ietfa.amsl.com> X-Forwarded-Message-Id: <20111031194200.18624.29794.idtracker@ietfa.amsl.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Subject: [ledbat] Fwd: New Version Notification - draft-ietf-ledbat-congestion-09.txt X-BeenThere: ledbat@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: janardhan.iyengar@fandm.edu List-Id: Mailing list of the LEDBAT WG List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 31 Oct 2011 19:54:43 -0000 Dear all, A new version of the LEDBAT draft (-09) is online now. We have hopefully taken in all the feedback through WGLC; please let us know if we've missed anything! Thanks, - jana -------- Original Message -------- Subject: New Version Notification - draft-ietf-ledbat-congestion-09.txt Date: Mon, 31 Oct 2011 12:42:00 -0700 From: internet-drafts@ietf.org To: ledbat-chairs@tools.ietf.org, draft-ietf-ledbat-congestion@tools.ietf.org, wes@mti-systems.com New version (-09) has been submitted for draft-ietf-ledbat-congestion-09.txt. http://www.ietf.org/internet-drafts/draft-ietf-ledbat-congestion-09.txt Diff from previous version: http://tools.ietf.org/rfcdiff?url2=draft-ietf-ledbat-congestion-09 IETF Secretariat.