The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2004-Feb> msg00035



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Fwd: New draft on Inter-area/AS MPLS TE: draft-vasseur-ayyangar-inter-area-AS-TE-00.txt

  • From: Jean Philippe Vasseur <jvasseur@cisco.com>
  • Date: Fri, 06 Feb 2004 13:41:04 -0500

Just a short email to mention your comments on http://www.ietf.org/internet-drafts/draft-vasseur-ccamp-inter-area-as-te-00.txt
are welcome on the CCAMP mailing list.

JP.

Date: Wed, 04 Feb 2004 17:06:33 -0500
To: ccamp@ops.ietf.org
From: Jean Philippe Vasseur <jvasseur@cisco.com>
Subject: New draft on Inter-area/AS MPLS TE: draft-vasseur-ayyangar-inter-area-AS-TE-00.txt
Cc: arthi Ayyangar <arthi@juniper.net>, adrian@olddog.co.uk, kireeti@juniper.net

Hi,

During IETF-58, two drafts related to Inter-area and Inter-AS MPLS TE were presented:
        - draft-vasseur-inter-AS-TE (JP Vasseur / Raymond Zhang)
        - draft-ayyangar-inter-region-te (Arthi Ayyangar)
It was agreed with the WG to merge the two drafts which should cover solutions for both inter-area and inter-AS MPLS TE and for both packet and non-packet TE LSPs. Here is the result of this merge.

In a nutshell, the draft proposes three TE LSPs types (Contiguous, Stitched and Nested) and two inter-area/AS TE path computation methods (Per-area/AS, distributed PCEs). We tried to also describe the applicability of each TE LSP type and path computation method as well as their respective pros and cons with regards to several aspects (path optimality, reoptimization, diverse path computation, DS-TE, hierarchy, policy control, ...).

Just a side note regarding the terminology: the comment was made several times that the PCS terminology was confusing and may be interpreted as "off-line". Hence this has been changed for PCE (Path Computation Element), where the PCE may be an ABR or an ASBR. This applies to one of the distributed path computation method described in this draft.

Several SPs expressed an interest in this draft.

Comments are obviously more than welcome !

We'd like to specifically thank Adrian Farrel for his contribution to this merged draft.

JP.