Re: [Pce] Charter and Multi-Domain

[JP Vasseur <jvasseur at cisco.com>]

Hi Dan,

*co-chair hat off*

On Nov 3, 2009, at 9:08 PM, Daniel King wrote:

Hi Julien, all.

Good topic.

When we starting to work on the H-PCE architecture
(draft-king-pce-hierarchy-fwk-xx) it was in response to perceived scaling
concerns of BRPC, if the sequence of domains was not pre-defined. Our view
is that solving the domain routing problem for as few as six interconnected
domains could be really hard if BRPC is used without a pre-defined domain
sequence.  

Two comments:

1) I personally think that the proposed mechanism is technically *good*

2) That being said, this raises the question on whether we need such mechanism considering 

the relatively small scale of current deployments and I would welcome any evidence of lack of

scalability of BRPC in small scale network deployments. 

So, I think our I-D is within the current WG charter. I'd actually go a step
further and say that our I-D is not aimed at a larger set of domains at this
stage.

Lastly, I do welcome the WG discussing extending the scope of the charter.
Perhaps it may be worth defining what we mean by a "larger groups of
domains", it's certainly >2 and <Internet. Would up to 20 domains be a good
reference point when defining/discussing a large group of domains?

Well, as Adrian knows, I definitely do not like specifying hard numbers when it comes to defining

IP technologies since I do believe that one of the reasons of the success of the IP is precisely because

no numbers were specified in the original specifications. But yes 20 seems a good reference point.

Cheers.

JP.

Br, Dan.

-----Original Message-----
From: pce-bounces at ietf.org [mailto:pce-bounces at ietf.org] On Behalf Of
julien.meuric at orange-ftgroup.com
Sent: 03 November 2009 16:49
To: pce at ietf.org
Subject: [Pce] Charter and Multi-Domain

Hi PCE members.

As you may (SHOULD ;-) ) have noticed in our agenda for Hiroshima, Daniel
King will present his I-D named "The Application of the PCE Architecture to
the Determination of a Sequence of Domains in MPLS & GMPLS" (introduced
during the Stockholm meeting). Nonetheless our current charter soberly says:
"The PCE WG will work on application of this model within a single Domain
or within a small group of domains (where a domain is a layer, IGP area
or Autonomous System with limited visibility from the head-end LSR). At
this time, applying this model to large groups of domains such as the
Internet is not thought to be possible, and the PCE WG will not spend
energy on that topic."

In other words, we are flirting with our charter border and the scope of
some drafts depends on it. In this context, the chairs would like to get a
feedback from the WG on evolving towards larger groups of domains.
Does the WG think we are now ready to follow this path? Do you have
operational requirements and/or concerns in mind in terms of applicability,
order of magnitude, implementation...? Comments are very welcome.

You are probably aware that we have passed over a milestone about
rechartering: beyond the upcoming meeting, your feedback will be useful
input for the envisioned charter update (targetted for the end of the year).

Many thanks

JP & Julien
_______________________________________________
Pce mailing list
Pce at ietf.org
https://www.ietf.org/mailman/listinfo/pce