Re: Please read and comment on GMPLS P&R DT documents

["Mario Molinari" <Mario.Molinari@marconi.com>]

Hi P&R Design Team,

These are my comments on document
draft-mannie-gmpls-recovery-terminology-00.txt.

In paragraph 4.6 (Recovery types), sub-paragraphs D and E,
you state that "all these (working and recovery) LSPs/spans
must start and end at the same nodes".
In an end-to-end restoration scheme, couldn't we have LSP's
with different end nodes sharing the same restoration
resources?

For example, if restoration were pre-planned, the restoration
path computation process might select a route sharing resources
with the restoration route of other LSP's (possibly terminating
on different nodes), as long as it could guarantee that there
were no single risk affecting more than one of the LSP's using
the same restoration resource.


In paragraph 4.13 (External commands), I don't agree to some
of the terms used.

I would rename "Lockout of normal traffic" to "Forced switch
to working" and "Forced switch for normal traffic" to "Forced
switch to recovery".
I'd redefine "Lockout of normal traffic" as in the following.
"A configuration action initiated externally that results in
the normal LSP/span being temporarily unavailable to transport
traffic".
I'd also rename "Manual switch for normal traffic" into "Manual
switch to recovery", and add a "Manual switch to working", which
would be defined as in the following.
"A switch action initiated externally that switches normal
traffic (back) to the working LSP/span, unless a fault condition
exists on such LSPs/span or an equal or higher priority switch
command is in effect."

As a more general statement, it seems to me that this section
borrows from G.841 terminology, but introduces some changes
(like the ones for which I proposed an amendment). Which is the
reason for such changes, apart of the use of "recovery" instead
of "protection", which is obviously justified by the context?


Mario Molinari




Sudheer Dharanikota <sudheer@nayna.com>@ops.ietf.org on 13/03/2002 18.27.51

Sent by:  owner-ccamp@ops.ietf.org


To:   CCAMP WG <ccamp@ops.ietf.org>
cc:

Subject:  Please read and comment on GMPLS P&R DT documents


Hi Folks:

Please note that the "GMPLS Protection and Restoration
Design Team"
has produced the following two documents (1st version)...

Recovery (Protection and Restoration) Terminology for GMPLS
http://search.ietf.org/internet-drafts/draft-mannie-gmpls-recovery-terminology-00.txt


Analysis for GMPLS-based Recovery Mechanisms(including
Protection and Restoration)
http://www.cs.odu.edu/~sudheer/draft-design-team-gmpls-recovery-analysis-00.txt


We request you to read and comment on them as quickly as you
can.

Regards,

GMPLS P&R Design Team