Resource Guarantee for SRv6 Policies

Abstract

This document defines a new SRv6 Endpoint behavior which can be used to associate with a set of network resource partition (e.g. bandwidth, buffer and queue resources ) Programming, called End.NRP. By using the End.NRP SID to build its segment list, the SRv6 policy has the capability to program network resources and achieve strict SLA guarantees.¶

1. Introduction

The concept of Network Resource Partition is introduced in [RFC9543]. A Network Resource Partition (NRP) is a set of network resources that are allocated from the underlay network to carry a specific set of network traffic and meet the required SLOs and SLEs.¶

Segment Routing (SR) [RFC8402] leverages the source routing paradigm. An ingress node steers a packet through an ordered list of instructions, called "segments". Each one of these instructions represents a function to be called at a specific location in the network. A function is locally defined on the node where it is executed and may range from simply moving forward in the segment list to any complex user-defined behavior.¶

SR Policy is an ordered list of segments (i.e. instructions) that represent a source-routed policy. The packets steered into an SR Policy have an ordered list of segments associated with that SR Policy.¶

Since the SRv6 Endpoint behavior defined in [RFC8986] are not associated with a set of network resource partition of the interface for slices/slice aggregate(e.g.End.X just forwards to an endpoint with cross-connect to a 'layer-3 adjacency' or L2 bundles). Therefore, SRv6 policies can't achieve strict SLA guarantees.¶

[I-D.ietf-spring-resource-aware-segments]extends the SR paradigm by associating SIDs with network resource attributes. On the basis of [I-D.ietf-spring-resource-aware-segments], this document defines a new SRv6 Endpoint behavior which can be used to associate with a set of network resource partition (e.g. bandwidth, buffer and queue resources ) Programming, called End.NRP. By using the End.NRP SID to build its segment list , the SRv6 policy has the capability to program network resources and achieve strict SLA guarantees.¶

2. End.NRP Behavior

This section defines a new SRv6 Endpoint behavior which can be used to associate with a set of snetwork resource partition (e.g. bandwidth, buffer and queue resources ) Programming, called End.NRP. The End.NRP is a variant of the End.X behavior defined in [RFC8986].¶

Any SID instance of End.NRP behavior is associated with two sets: J1 and J2.¶

J1: one or more L3 adjacencies or L2 bundles
J2: NRP of J1

When N receives a packet destined to S and S is a local End.NRP SID, the line S15 from the End processing defined in [RFC8986] is replaced by the following:¶

S15. Submit the packet to the IPv6 module for transmission to
     the new destination via a member of J1, using the NRP
     identified by J2

This End.NRP SID SHOULD support the Penultimate Segment Pop (PSP) of the SRH, Ultimate Segment Pop (USP) of the SRH, and Ultimate Segment Decapsulation (USD) flavors defined in [RFC8986] either individually or in combinations. The SRH processing of the End.NRP behavior with PSP, USP, and USD is the same as [RFC8986].¶

This End.NRP SIDs can be allocated either by a centralized network controller or by the network nodes, and the End.NRP behavior MAY be announced using IGP or BGP-LS. The detailed protocol extension will be described in a separate document.¶

3. Use Cases for End.NRP behavior

This section describes possible procedures for the End.NRP behavior.¶

A group of End.NRP SIDs SHOULD be allocated for the set of network resources associated with the SRv6 Policies, so that different End.NRP SIDs SHOULD be used to steer service traffic into different set of link resources (e.g. bandwidth, buffer and queue resources) in packet forwarding.¶

Below is the possible procedures:¶

• The controller get the topology information, calculate the SR Policy path based on SLA.¶

• The controller cooperates with the network nodes to complete resource reservation and the End.NRP SID allocation along the SR Policy path.¶

• The controller use the End.NRP SID to build the SID list for the explicit path.¶

• Then the controller inform the headend the resource guaranteed path by various means including: via BGP [I-D.ietf-idr-sr-policy-safi], configuration or PCEP [RFC8664] [I-D.ietf-pce-segment-routing-policy-cp].¶

• 
  SRv6 Policy1 from PE1 to PE2
  segment list <End.NRP11, End.NRP12, End.NRP13>
  
  SRv6 Policy2 from PE1 to PE2
  segment list <End.NRP21, End.NRP22, End.NRP23>
  
                         End.NRP12
                         End.NRP22
                        * * * * * *
                      P1-----------P2
                    */*|* * * * * * |\ * *
                   */* |            | \ * *
  CE11-----   End.NRP11|            |End.NRP13  ------ CE12
           \  End.NRP21|            |End.NRP23 /
            \   */ *   |            |    \* * /
             PE1 \     |            |      PE2
            /     \    |            |    /    \
           /       \   |            |   /      \------ CE2
  CE21------        \  |            |  /
                     \ |            | /
                      P3------------P4
  

  ¶

Figure 1 shows an example for the End.NRP behavior.¶

As shown in Figure 1, there are two customers with different leased line requirements from PE1 to PE2:¶

• leased line1 : 1G BandWidth with strict SLA guarantee.¶

• leased line2 : 2G BandWidth with strict SLA guarantee.¶

Below is the possible procedures:¶

• The controller get the topology information, calculate the SRv6 Policy 1 and SRv6 Policy 2 based on SLA.¶

• The controller cooperates with the network nodes to complete resource reservation and the End.NRP SID allocation along the SRv6 Policy1 and SRv6 Policy 2. Taking the interface PE1-P1 of SRv6 node PE1 along the SRv6 Policy 1 as an example, two different NRPs(e.g. two dedicated queues) are partitioned from the network resources of the physical link PE1-P1 (GE1/0/0).¶

  • The NRP(Queue1:1G BW)of link PE1- P1 is reserved and associated with End.NRP11.¶

  • The NRP(Queue2:2G BW)of link PE1- P1 is reserved and associated with End.NRP21.¶

• The controller uses the End.NRP11, End.NRP12, and End.NRP13 to build the SID list for the SRv6 Policy1 and use the End.NRP21, End.NRP22, and End.NRP23 to build the SID list for the SRv6 Policy 2.¶

• Then the controller inform the headend the segment list of SRv6 Policy 1 and the segment list of SRv6 Policy 2 by various means including: via BGP [I-D.ietf-idr-sr-policy-safi], configuration or PCEP [RFC8664] [I-D.ietf-pce-segment-routing-policy-cp].¶

The traffic from customer1 and customer2 will be forwarded to PE2 through the NRPs previously reserved for each hop link on the path of SRv6 Policy1 and SRv6 Policy2 respectively, thus Customer 1 and Customer 2 are provided with end-to-end 1G bandwidth resources and 2G bandwidth resources respectively, and private line services are guaranteed by strict SLAs.¶

4. Security Considerations

The security requirements and mechanisms described in [RFC8402], [RFC8754] and [RFC8986] also apply to this document.¶

This document introduces a new SRv6 Endpoint behavior for implementation on the nodes support network resource partition in the network. As such, this document does not introduce any new security considerations.¶

5. Implementation Status

[Note to the RFC Editor - remove this section before publication, as well as remove the reference to [RFC7942]].¶

This section records the status of known implementations of the protocol defined by this specification at the time of posting of this Internet-Draft, and is based on a proposal described in [RFC7942]. The description of implementations in this section is intended to assist the IETF in its decision processes in progressing drafts to RFCs. Please note that the listing of any individual implementation here does not imply endorsement by the IETF. Furthermore, no effort has been spent to verify the information presented here that was supplied by IETF contributors. This is not intended as, and must not be construed to be, a catalog of available implementations or their features. Readers are advised to note that other implementations may exist.¶

According to [RFC7942], "this will allow reviewers and working groups to assign due consideration to documents that have the benefit of running code, which may serve as evidence of valuable experimentation and feedback that have made the implemented protocols more mature. It is up to the individual working groups to use this information as they see fit".¶

8. References