Dear PCErs,

FYI, a new version of draft-zhang-pce-hierarchy-extensions has been uploaded.
This document defines PCEP extensions for H-PCE (see below).

It is a minor update, and, most importantly, a lot of open issues remain.
In this sense, your comments / suggestions are welcome.

The current draft already proposes some protocol extensions / encodings to the
high-level procedures detailed in the companion framework document. Recent
exchanges seem to imply that there is room to revisit requirements and
procedures,including

* (Parent) TED management: it can be argued that the framework document relies
  on basic domain connectivity, precluding TE aggregation mechanisms. This
  seems to be a topic of debate. It has been mentioned in private conversations
  that it should be a decision of the network operator (based on policies or
  whatever criterion) to decide the level of information exchange allowed. It
  clearly seems to depend on the actual scenario (e.g. multi-area WSON vs
  Inter-AS/carriers)

* In any case, decoupling TED management from path computation has been a
  design criterion within the PCE architecture and this should be applied also
  to the solutions document. Several mails have hinted the idea of (separate?)
  approaches to (hierarchical) TED management (cfr. Olivier/Oscar mails) and
  leaving TED management out of scope (including wrapping TED-related messages
  out of PCEP / PCNtf). It may be difficult if both functions (path computation
  and TED management) are slightly coupled (border node identification,
  endpoint localization and topology aggregation for domain sequence selection,
  to name a few in which an IGP-based TED may not be sufficient)

* Domain representation: the role of domains and domain identifiers is clearly
  important. We should align with e.g. draft-dhruv on domain sequence, avoid
  new encodings for domains which could re-use Route sub-objects, etc.

* Alternative approaches (Dhruv) suggest to keep the Parent-PCE’s topology
  graph free of BNs (Boundary Nodes) and inter-AS TE link; it being composed
  only of neighbor domain adjacency.

* OF codes: new requirements arise regarding OF codes, including the
  consideration of actual OF codes (e.g. minimize domain crossing) and
  policies affecting them (e.g. do not allow domain re-enter).  Additionally,
  there seems to be a need to be able to specify the OF codes to apply at both
  levels, not only at the parent level but also the child's (i.e. intra-domain
  level)

* Reachability: the current fwk/pcep drafts rely on polling to locate
  endpoints. This does not preclude some form of caching, but alternative
  solutions based on notifications/announcements of endpoint (prefixes) have
  been mentioned.

* Exclusions: exclusions need to take into account domains.

* Other?...


Dan mentioned that discussion and agreement of the above, will help create
requirements that will allow us define the Protocol extensions, including PCEP
extensions, encoding, error handling and manageability. For this, Dan plans to
leave the Framework document open in case new requirements arise.


Best regards,

Ramon, on  behalf of the draft co-authors



-------------------------------------------------------------------------
New Version Notification for draft-zhang-pce-hierarchy-extensions-01.txt

A new version of I-D, draft-zhang-pce-hierarchy-extensions-01.txt has been 
successfully submitted by Fatai Zhang and posted to the IETF repository.

Filename:        draft-zhang-pce-hierarchy-extensions
Revision:        01
Title:           Extensions to Path Computation Element Communication Protocol 
(PCEP) for Hierarchical Path Computation Elements (PCE)
Creation date:   2011-10-13
WG ID:           Individual Submission
Number of pages: 19

Abstract:
   The hierarchical Path Computation Element (PCE) architecture, defined
   in the companion framework document [I-D.ietf-pce-hierarchy-fwk],
   allows the selection of an optimum domain sequence and the optimum
   end-to-end path, to be derived through the use of a hierarchical
   relationship between domains.

   This document defines the Path Computation Element Protocol (PCEP)
   extensions for the purpose of implementing hierarchical PCE
   procedures which are described the aforementioned document.

--
Ramon Casellas, Ph.D.
Research Associate - Optical Networking Area -- http://wikiona.cttc.es
CTTC - Centre Tecnològic de Telecomunicacions de Catalunya, PMT Ed B4
Av. Carl Friedrich Gauss, 7 - 08860 Castelldefels (Barcelona) - Spain
Tel.: +34 93 645 29 00 -- Fax. +34 93 645 29 01

_______________________________________________
Pce mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/pce

Reply via email to