1. Slide 1 MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks MPLS-TP Linear Protection Nurit Sprecher Yaacov Weingarten

2. Slide 2MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks General Linear protection provides a fast and simple protection switching mechanism. It provides a clear indication of the protection status. Linear protection fits best in mesh networks. Terminology based on RFC 4427

3. Slide 3MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Elements of Protection Linear protection may be used to protect against a failure of one of the following point-to-point entities (elements of protection): – End-to-end LSP (between a pair of LERs) – End-to-end PW (between a pair of T-PEs) – Tandem Connection (TC): ▪ LSP segment: Between an LER and any LSR of a given LSP Between any two LSRs of a given LSP ▪ One or more PW segments: Between a T-PE and any S-PE of a given MS-PW Between any two S-PEs of a given MS-PW. It can span several PW segments. – Span *** Point-to-multipoint is for further study.

4. Slide 4MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Architecture Linear protection operates in the context of a Protection Domain. A Protection Domain is composed of the following architectural elements: – A pair of end points (East and West ) which reside at the boundary of the Protection Domain. ▪ In each transmission direction one of the end points is referred to as a source and the other one is referred to as a sink. – A Protection Group which consists of a Working entity and a Protection entity ▪ A Working entity can be of any type of element of protection (an end-to-end LSP or an end-to-end PW or Tandem Connection or a Span) ▪ A Protection entity is a dedicated entity whose function is to protect against a failure of the Working entity. Linear protection is a fully allocated survivability mechanism. The route and the bandwidth of the Protection entity is pre-provisioned. – Protection switching SHOULD be guaranteed, thus the Working and the Protection entities SHOULD be disjoint entities, i.e. the physical routes of the Working and the Protection entities SHOULD have complete physical diversity.

5. Slide 5MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Architectural Model Protection Domain Working Entity Protection Entity WestEast Protection Switching Controller Protection Group

6. Slide 6MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Functional Architecture In each transmission direction, traffic is bridged at the source end point onto the appropriate entity, and is selected from the appropriate entity at the sink end point. Protection switching occurs at the protection switching controllers which reside at the boundaries of the Protection Domain.

7. Slide 7MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Switching Types Bidirectional protection switching: – In the event of failure, traffic in both directions is switched to the standby entity (even when the fault is unidirectional). – An attempt should be made to coordinate the two ends (so that both have the same bridge and selector settings). Unidirectional protection switching: – Only traffic transmitted in the affected direction is switched to the standby entity. The selectors on both sides are independent.

8. Slide 8MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Bidirectional 1:1 Protection Switching Traffic is transmitted either onto the Working or onto the Protection entity (using a selector bridge at the source). The selector at the sink selects the entity which carries the traffic. A mechanism is needed to force the selectors in both directions to select the same entity. When there is no failure, the standby entity may carry extra traffic (which may be preempted).

9. Slide 9MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Bidirectional 1:1 Protection Switching Protection Domain Working Entity Protection Entity Selector Permanent Bridge Normal Condition Protection Condition Failure Condition Working Entity Protection Entity Selector Permanent Bridge Optional Extra Traffic (preempted)

10. Slide 10MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Unidirectional 1+1 Protection Switching Traffic is copied and bridged at the source to both Working and Protection entities (with a permanent bridge at the source). The selection is made at the sink based on a number of predetermined criteria, such as defect indication, management request, etc. There is no need for a mechanism to coordinate the two ends.

11. Slide 11MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Unidirectional 1+1 Protection Switching Protection Domain Working Entity Protection Entity Selector Permanent Bridge Normal Condition Protection Condition Failure Condition Working Entity Protection Entity Selector Permanent Bridge

12. Slide 12MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Revertive/Non-Revertive Operation In revertive operation, when the failure condition is eliminated, the protected traffic is switched back onto the Working entity. – To verify that the network has stabilized, and to avoid frequent switching in case of intermittent failures, traffic is not switched back to the Working entity before the Wait-to-Restore (WTR) timer has expired. In non-revertive operation, when the failure condition is eliminated, the protected traffic remains on the Protection entity. Revertive/non-revertive operations are provided as network operator options.

13. Slide 13MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Protection Switching Trigger Actions Protection switching may be performed when: – A failure condition (‘failed’ or ‘degraded’) is declared on the active entity and is not declared on the standby entity. – An external operator command is received (e.g. ‘Forced Switch’, ‘Manual Switch’). – A request to switch over is received from the far end - relevant in case of bidirectional 1:1 protection switching only! Priorities should be set between the various triggers.

14. Slide 14MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Protection Switching Trigger Actions Failure Condition Requires Continuity Check (CC) monitoring of both Working and Protection entities: – The protection switching controller does not care which monitoring method is used, as long as it can be given information for the Working and the Protection entities (‘ok’, ‘failed’ or ‘degraded’). – For protection switching, it is common to run a CC every 3.33ms*. In the absence of three consecutive CC messages, a ‘failed’ condition is declared. ▪ Note that Protection switch completion time excludes the detection time necessary to initiate the protection switch, and the hold-off time. – In order to monitor the Working and the Protection entities, an OAM Maintenance Entity should be defined for each of the entities. OAM information should be provided as input to the protection switching controllers.

15. Slide 15MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Protection Switching Trigger Actions External Administrative Commands The following manual commands may be applied to a protection group: Clear – results in clearing all the active external administrative commands Lockout of Protection – fixes the selector position to the Working entity. Results in the Protection entity being temporarily unavailable to transport traffic (either normal or extra traffic). This command actually disables the protection group. Force protection action – forces a switch of the traffic to the Protection entity Manual switch to the Working entity – switches the traffic to the Working entity unless a ‘failure’ condition exists on the Working entity Manual Switch to the Protection entity – switches the traffic to the Protection entity unless a ‘failure’ condition exists on the Protection entity *** The commands are listed in descending order of priority.

16. Slide 16MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Protection Switching Trigger Actions Peer State Coordination (PSC) In bidirectional 1:1 protection switching, an attempt is made to coordinate the protection switching state between both ends of the Protection Domain. This includes also the signaling of: – Unidirectional failure condition – External administrative requests PSC is also used to detect mismatches between the provisioned protection switching configuration and the two ends of a Protection Domain. *** Note: If there is no requirement to signal external administrative requests and to check configuration consistency at both ends, the RDI (Remote Defect Indication) tool can be used instead of PSC to signal a unidirectional fault condition.

17. Slide 17MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Hold-off Timer A hold-off timer is enabled to coordinate the timing of protection switching at multiple layers or across cascaded protected domains. – Detection of a failure condition is not reported to the protection switching controller until the hold-off timer reaches zero. When the hold-off timer expires, and a failure condition still exists, the condition is reported to the protection switching controller. This either allows the server layer to fix the problem before protection switching occurs in the MPLS-TP layer, or, alternatively, it allows the upstream protected domain to switch before a downstream domain.

18. Slide 18MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Management Plane The management plane may be used to configure the protection domain and the following elements: – Protection switching controllers – Protection group: Working and Protection entities – Protection type (i.e. bi-directional 1:1 linear protection or unidirectional 1+1 protection switching) – Additional parameters, such as a hold-off timer, revertive/non-revertive operation, etc.) The management plane may initiate manual control of the protection switching function. The management plane should be able to monitor the protection switching status of the protected domain.

19. Slide 19 MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks

20. Slide 20 MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Backup Slides

21. Slide 21MPLS-TP Linear Protection / Author / 25-8-2008 RTP IE Fixed CET I insert classification level © Nokia Siemens Networks Priority of Protection Switching Trigger Actions Local RequestOrder of Priority ClearHighest Lockout of Protection| ‘Failed’ condition on Protection| Forced Switch| ‘Failed’ condition on Working| ‘Degraded’ condition on Working| Manual Switch| Wait to Restore| No TriggerLowest