IETF-87 pce agenda


1. Introduction
2.1.1. Requirements for RWA
2.1.2. PCEP Extensions for RWA
2.2. PCEP Extensions for Enhanced TE Parameters
3.1. Current Issues with Existing RBNF Notation
3.2. PCEP Extensions for Hierarchical PCE
3.3. Path-Key MIB
3.4. PCE Database Requirements
3.5. DNS-Based PCE Discovery
4.1. Stateful PCE Applicability
4.2. PCEP Extensions for Stateful PCE and PCE-Initiated LSPs
4.3. PCE-Initiated GMPLS LSP
4.4. LSP Database Synchronization
5.1. PCEP Extensions for Automatic Bandwidth Adjustement
5.2. PCEP Extensions for Segment Routing
5.3. Stateful PCE Extensions for Data Plane Switchover and Balancing
5.4. Make-Before-Break MPLS-TE LSP Restoration and Reoptimization
5.5.Discovery of Stateful Capability

Session 2013-07-31 1510-1610: Tiergarten 1/2 - Audio stream - pce chatroom
Session 2013-07-31 1620-1720: Tiergarten 1/2 - Audio stream - pce chatroom


PCE Working Group Meeting - Wednesday, July 31, 2013; 3:10-5:20 PM

1. Introduction

1.1. Administrivia, Agenda Bashing (chairs, 5 min) [5/130]

- No agenda changes.

1.2. WG Status (chairs, 15 min) [20]


- Several comments and discuss on the list.
- All issues closed to being resolved new document to be submitted shortly.
- Still need to resolve BW discussion, looks to be an issue with semantics and the discussion will be move to the list.

- Two encoding options exist in the draft, please review and comment on the list and/or to the authors.

PCEP MIB Documents
- PCEP MIB documents progressed well and stable, ready for Last Call.
- PCEP Discovery MIB updated.
- PCEP TC MIB document is no longer used.

PCE Management
- Management of PCE raised by Operations AD. Adrian Farrel will be creating a design team for Network PCE Management. He is looking for volunteers.

2. WG I-Ds

2.1. Requirements and PCEP Extensions for RWA (Young Lee, 5 min) [25]



- Co-chairs highlighted wavelength constraint needs to be more specific and requested that the authors confirm via the working group list that what is proposed is suitable.

2.2. PCEP Extensions for Enhanced TE Parameters (Dhruv Dhody, 5 min) [30]


- No questions.

3. Known Topics and Discussed I-Ds

3.1. Current Issues with Existing RBNF Notation (Ramon Casellas, 5 min) [35]


- Co-chairs highlighted interesting nature of the work and encouraging working group to review and comment.

3.2. PCEP Extensions for Hierarchical PCE (Dan King, 5 min) [40]


- Co-chairs conducted a Poll, good support from the room. Poll will be taken to the list.

3.3. Path-Key MIB (Dhruv Dhody, 5 min) [45]


- Need discussion on if SNMP will continue to be a viable technology for management of PCE.
- Co-chairs requested that this be one of the questions of the PCE management design team.

3.4. PCE Database Requirements (Oscar Gonzalez de Dios, 5 min) [50]


- Comments that BGP-LS is not discussed in detail in the document, and LS-DB and TED concepts seem to be combined and they should be shown as separate entities with dependencies.

3.5. DNS-Based PCE Discovery (Qin Wu, 5 min) [55]


- Comment regarding the need for PCE discovery and is it actually required? Co-chairs highlighted that IGP discovery already exists, and this DNS discovery mechanism is being proposed for use cases and open issues.
- Comment from Operator that this is applicable to inter-domain and H-PCE.
- Further comment that Zookeeper/Curator might suitable for discovery functions.
- Co-chairs request that authors discuss the draft with DNS experts to verify suitability.

4. Stateful PCE

4.1. Stateful PCE Applicability (Ina Minei, 5 min) [60]


- Co-chairs conducted a Poll, good support from the room. Poll will be taken to the list.

4.2. PCEP Extensions for Stateful PCE and PCE-Initiated LSPs (Ina Minei, 15 min) [75]



- Comment from authors for the WG adoption of the document.
- Co-chairs pointed out that overlap exists between MPLS and GMPLS drafts and we need to progress the discussion on the working group list.
- Authors highlighted that discussion on finding a way forward to progress work should be straightforward

4.3. PCE-Initiated GMPLS LSP (Victor Lopez, 5 min) [80]


- Authors have no issue with merging with draft-crabbe-pce-pce-initiated-lsp.
- Operator comment that having a single document with multiple options can cause issues when you only want support for a specific feature.
- Another Operator comment should support for a joint solution for instantiation solution development, with sections focused on the different uses (MPLS-TE and GMPLS).
- Co-chairs do not feel that merging and increasing the number of options in the draft for varying technologies is a major issue, but they would like to hear other options and feedback on the list.

4.4. LSP Database Synchronization (Xian Zhang, 15 min) [95]



- Comment on incremental synchronisation, timing should not preclude the PCE requesting a full synchronisation (verification, etc.).
- Suggestion to include optimisation in the stateful core draft. The authors of the core draft have concerns that the document is already large, and they would prefer to avoid adding new function.
- Co-chairs also expressed support for a single draft for stateful synchronisation timing and optimisation. This topic will be taken to the list for further discussion.
- Comment for second draft (draft-palle-pce-stateful-pce-lspdb-sync-00) suggestion that the PCCs could synchronise instead of PCEs. Co-chairs suggest this aspect also be moved to the list.

5. New I-Ds not discussed...

5.1. PCEP Extensions for Automatic Bandwidth Adjustment (Dhruv Dhody, 5 min) [100]


- Comment that this is not particularly versatile, and constrained to vendor implementation.
- Concern that this proposal does not scale and does not discuss the impact of carrying periodic information. Authors will include an impact analysis in the next version of the document.
- Authors highlighted that some of requirements (stateful auto-BW) in the I-D are driven by Operator use cases.
- Further questions (carrying additional parameters and configuration learning) will be submitted to the list.

5.2. PCEP Extensions for Segment Routing (Jan Medved, 10 min) [115]


- Suggestion for the I-D to support existing PCEP mechanisms. Response that the I-D is focused on function required/supported for segment routing.
- Co-chairs highlighted that this is a new IETF topic, PCE has a role in segment routing and feedback is requested on the WG mailing list.

5.3. Stateful PCE Extensions for Data Plane Switchover and Balancing (Yosuke Tanaka, 5 min) [120]


5.4. Make-Before-Break MPLS-TE LSP Restoration and Reoptimization (Yosuke Tanaka, 5 min) [125]


- No comments.

---------- Depending on time left... ---------- [130/130]

- Discovery of Stateful Capability (Jan Medved, 5 min)


- TLS for PCEP (Diego Lopez, 5 min) -> Also on KARP agenda


- Due to running out of time Co-chairs asked that authors of the previous two I-Ds send a work summary and requests for review to the list.