Cyril,

Thank you for the very thoughtful comments. Please see inline ###.

Ina

From: Margaria, Cyril (Coriant - DE/Munich) [mailto:[email protected]]
Sent: Monday, July 08, 2013 5:34 AM
To: Ina Minei; JP Vasseur (jvasseur); Julien Meuric; [email protected]
Subject: RE: Stateful PCE applicability

Hi,

Yes we need an applicability document.
draft-zhang-pce-stateful-pce-app is addressing this, but there are some parts 
that could be improved:

Section 4.3 :  This is a drawback of the stateful PCE, this could be stated as 
follows :

A staful PCE requires an LSP-DB synchronization, which cause an addition delay 
or synchronization issues, thus impacting negatively the survivability of a 
PCE. .

In my opinion a statfull PCE could mitigate that by acting as stateless until 
this synchronization has been done.
### I agree we should discuss on potential drawbacks. As you have seen, there 
are various proposals (new drafts) submitted to alleviate this.

Section 5: maybe describing some use case not solved by a stateful PCE would be 
useful,
or which additional constraints this add
### Discussion on constraints added by a stateful pce deployment is probably 
something we should consider adding, but I wonder if this shouldn't fit better 
in section 4 (e.g. discussion on state sync)
For instance having an active stateful add another controller in the network, 
it may not always sit well with existing NMS or network architecture, yet they 
would benefit from the passive stateful.
### Not sure what you mean, maybe you have specific text?

Section 5 : it would be usefull to indicate which scenario requires an active 
stateful,
For instance section several use cases can be solved using both, an active 
stateful can fix the problem afterwards, a passive stateful could solve it 
beforehand (if the planned services are known),

So cases (for instance 5.4.2 or 5.4.3) can be solved using passive stateful PCE 
only, which would not present the same implication for deployement.
### The draft doesn't go in a lot of discussion on active and passive, this was 
not a goal.  I can see the point you are making, will evaluate with the 
co-authors how to address the comment on sections 5.4.2 and 5.4.3 in the next 
version.

Mit freundlichen Grüßen / Best Regards
Cyril Margaria
From: [email protected]<mailto:[email protected]> 
[mailto:[email protected]] On Behalf Of Ina Minei
Sent: Friday, June 21, 2013 6:17 PM
To: JP Vasseur (jvasseur); Julien Meuric; [email protected]<mailto:[email protected]>
Subject: [Pce] Stateful PCE applicability

Dear chairs and working group,

In light of the recent working group re-charter which now includes stateful 
PCE, we wanted to hear the opinions of the group on

1.      the need for an applicability document for stateful PCE and

2.      whether draft-zhang-pce-stateful-pce-app satisfies this need, or any 
gaps it might have

Thank you,

Ina and Xian

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

Reply via email to