BGP SR Policy Extensions for Path Scheduling

Introduction Segment Routing (SR) policy is a set of candidate SR paths consisting of one or more segment lists and necessary path attributes. It enables instantiation of an ordered list of segments with a specific intent for traffic steering. specifies how BGP may be used to distribute SR Policy candidate paths. It introduces a BGP SAFI with new NLRI to advertise a candidate path of a Segment Routing (SR) Policy. introduces a set of use cases where the topology of the network changes predictably. The topology change may cause some of the paths invalid, and lead to path reselection or even recalculation. However, the reselection or recalculation takes a period of time, which will affect packet forwarding and cause problems such as packet disorder and packet loss. However, on a network with predictable topology changes, the ingress node knows future topology changes, it can schedule the forwarding paths in advance, and steer flows to different set of paths based on time to prevent packet forwarding from being affected by topology changes. In the scenario of SR-policy-based TE paths, the resource allocation efficiency is a big challenge. Some flows just last for a short time, but the TE paths resources for the flows are usually reserved for a long time and can not be used by other services. Therefore, the SR policy originator can generate a policy with time-limited paths and resources, the head node schedules these paths over time to improve the utilization of resources. This document proposes extensions to BGP SR Policy to indicate the enable time and disable time of each candidate path/SR list. The policy originator can deliver multiple paths with different valid time. The ingress node determines the current valid paths based on the arrival time of the packet and forwards along the path.

Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 when, and only when, they appear in all capitals, as shown here.

Motivation This section describes the use cases that may benefit from scheduled paths.

Time Variant Network Use Case introduces the time variant network use cases, the tidal network is one of the typical time variant network scenarios. In the tidal network, in which the traffic volume varies greatly at different time. In order to reduce the power consumption, some of the links and nodes may be shut down when the traffic is at a low level. In this scenario, the SR policy originator can generate a policy with several paths for different topology. The ingress node doesn’t need to wait for the advertisement of topology change and just changes the forwarding path based on the valid time of each path, the affection of topology change is minimized.

Resource utilization efficiency use case Traditionally, the usage and allocation of network resources, especially bandwidth, can be supported by a Network Management System (NMS) operation such as path pre-establishment. However, this may not provide efficient usage of network resources. The established paths may reserve the resources for a long time. During this period, the resources cannot be used by other services even when they are not used for transporting any service. In the scenario of SR-policy-based TE path, the resource allocation efficiency is also a problem. Some flows just last for a short period of time, but the TE paths resources for the flows are usually reserved for a long time and cannot be used by other services. In this scenario, the controller (originator of the SR policy) can calculate a path with a valid time based on the flow duration. When the TE path is invalid, the ingress node does not steer any packets to the path and releases the resources. Furthermore, the controller can use the resource released by the flow to plan TE paths for other flows that do not have overlap time, which can effectively improve the utilization of network resources.

Scheduling time information in SR Policy extends the stateful PCE Communication Protocol (PCEP) to enable Label Switched Path (LSP) path computation, activation, setup, and deletion based on scheduled time intervals for the LSP and the actual network resource usage in a centralized network environment so as to improve the resource utilization efficiency. Similar with , this document extends the SR Policy to enable paths scheduling. The NLRI defined in contains the SR Policy candidate path. The content of the SR Policy Candidate Path is encoded in the Tunnel Encapsulation Attribute defined in using a new Tunnel-Type called SR Policy Type with codepoint 15. The SR Policy encoding structure is as follows: Attributes: Tunnel Encapsulation Attribute (23) Tunnel Type: SR Policy (15) Binding SID SRv6 Binding SID Preference Priority Policy Name Policy Candidate Path Name Explicit NULL Label Policy (ENLP) Segment List Weight Segment Segment ... ... ]]> A candidate path includes multiple SR paths, each of which is specified by a segment list. The Scheduling time information can be applied to the candidate path, so that all the segment list of each condidate path have the same scheduling time. The new SR Policy encoding structure is expressed as below: Attributes: Tunnel Encapsulation Attribute (23) Tunnel Type: SR Policy (15) Binding SID SRv6 Binding SID Preference Priority Policy Name Policy Candidate Path Name Explicit NULL Label Policy (ENLP) Scheduling Time Information Segment List Weight Segment Segment ... ... ]]> The Schduling time information also can be applied to each segment list to indicate the valid time for each segment list, the new SR Policy encoding structure is expressed as below: Attributes: Tunnel Encapsulation Attribute (23) Tunnel Type: SR Policy (15) Binding SID SRv6 Binding SID Preference Priority Policy Name Policy Candidate Path Name Explicit NULL Label Policy (ENLP) Segment List Scheduling Time Information Weight Segment Segment ... ... ]]>

Scheduling time information Sub-TLV The Scheduling time information sub-TLV has two formats according to the change regularity of network topology. They are Aperiodic Scheduling Time Information (ASTI) sub-TLV and Periodic Scheduling Time Information (PSTI) sub-TLV.

ASTI sub-TLV The ASTI sub-TLV indicates one or more valid time (each includes the enable time and disable time) of one or more SR paths. The format of ASTI sub-TLV is shown as follows:

ASTI sub-TLV Type: TBD1 Length: the size of the value field in octets. Group Number: indicates the number of information groups, each information group has fields of Index, Flags, Enable Time and Disable time. Index: the number used to identify specific information group. This filed will be generated by the originate router. Enable Time: the time in seconds indicates when the path(s) is(are) enabled. Disable Time: the time in seconds indicates when the path(s) is(are) disabled Variable: one or more groups of time information (A information group is composed of Index, Flags, Enable Time and Disable time) may be included in one ASTI sub-TLV.

PSTI sub-TLV The PSTI sub-TLV indicates the enable time, disable time and period of one or more paths. The format of PSTI sub-TLV is shown as follows:

PSTI sub-TLV Type: TBD2 Length: the size of the value field in octets. Group Number: indicates the number of information groups, each information group has fields of Index, Flags, Period, Enable Time and Disable time. Index: the number used to identify specific information group. This filed will be generated by the originate router. Period: the time in seconds between the enable time of one repetition and the enable time of the next repetition. Enable Time: the time in seconds indicates when the path(s) is(are) enabled. Disable Time: the time in seconds indicates when the path(s) is(are) disabled Variable: one or more groups of time information(A information group is composed of Index, Flags, Period, Enable Time and Disable time) may be included in one PSTI sub-TLV.

Procedures When a SR Policy head node receives a SR Policy with scheduling time information, the head node will parse the SR Policy and save the scheduling time information locally. When a data packet arrives, the head node will steer it to a specific SR Policy by color or other means. Within a specific SR Policy, there are two ways for the head node to determine the final forwarding SR path: Option 1: A valid SR path is dynamically determined based on the packet arrival time whenever a packet arrives. Option 2: One or more valid paths are selected for the SR policy and one or more timers are set based on the path disable time. When the timer expires, packets steered to this SR policy are switched to another path.

Security Considerations These extensions to BGP SR Policy do not add any new security issues to the existing protocol.

IANA Considerations This document defines two new sub-TLV in the registry "BGP Tunnel Encapsulation Attribute sub-TLVs" to be assigned by IANA:

Value	Description	Reference
TBD1	Aperiodic Scheduling Time Information (ASTI) sub-TLV	This document
TBD2	Periodic Scheduling Time Information (PSTI) sub-TLV	This document