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

["Richard Rabbat" <rabbat@fla.fujitsu.com>]

Hi Design Team,

The following is the list of comments that I have on document:
draft-design-team-gmpls-recovery-analysis-00.txt

Comments include both substance as well as minor text corrections.

Page 2, line 53. 
- and (vertical hierarchy). Terminology used in this document is
- conforms to the one defined in [CCAMP-TERM].
+ and layering (vertical hierarchy). Terminology used in this document
+ conforms to the one defined in [CCAMP-TERM].

Page 2, line 41.
 -Signal Quality: monitors the performance of a signal. If
  the performance falls below a certain threshold a defect -
  excessive errors (EXC) or degraded signal (DEG) - is detected.

This seems to imply that signal performance monitoring only applies to
EXC and DEG. Changing the word to indicate that this is an example may
be better.

Actually, if this whole section is a re-write of SONET/SDH transport
plane capability, maybe just referencing the appropriate document makes
more sense to avoid duplication and possibly future changes in the
future from having to be constantly updated in documents.

Page 3, line 48.
-     comparing it with the expected identifier at the sink. Example,
-     the payload signal label (and the corresponding payload signal
-     mismatch detection).
+     comparing it with the expected identifier at the sink. An example
+     of payload types is the payload signal label (and the
corresponding
+     payload signal mismatch detection).

Page 3, line 54.
-   the next recovery phase) results in either:
+   the next recovery phase) result in either: 

Page 4, first paragraph.
   - Signal Fail (SF): A signal indicating the associated data has
     failed in the sense that a signal interrupting near-end defect
     condition (not being the degraded defect) is active.

Do you mean?
   - Signal Fail (SF): A signal indicating that the associated data has
     failed. This is a signal indicating that a near-end defect
     condition is active (as opposed to a degraded defect).

Page 4, line 16.
   ... In such cases, failure
   detection may be assisted by out-of-band communication, such as the
   one considered in [LMP-WDM] protocol which provides communication
   between the PXC and the OLS.

I think this should be more forcefully described. What conditions allow
for such assistance? How does the interaction between the 2 schemes
work, and what takes precedence?

Page 6, line 17.

"Too much" and "too few". It will be hard to quantify what is meant by
too much and too little information, since we do not currently discusss
information overload. Maybe we should wait for some protocol on fault
notification for protection to quantify the amount of information that
will be passed.

Page 6, line 39.

-     node) implies a consistent decisions based on the conditions for
+     node) implies that a consistent decision based on the conditions
for

Page 8, figure.

Maybe cut the figures into 2 separate figures and explain the axis for
the second figure.

Page 8, line 41
-   these operations are mostly planning time (using network planning
+   these operations are run mostly during planning time (using network
planning

Page 9, line 9
-   There are many mechanisms available allowing overbook the recovery
-   There are many mechanisms available allowing the overbooking of the
recovery

Page 9, line 15
-   best effort (recovered only if the resources are available).
-   unprotected (recovered only if the resources are available).  
It can be recovered if some restoration technique is applied.

Page 9, line 47
   Note: performance between the two approaches differs only by the
   time required for route computation and computation (and selection).

What is meant by the second computation?

---
I'll send an email later with the rest of the document, but I hope these
comments are helpful.

Richard
--
Richard Rabbat, Ph.D.
Fujitsu Laboratories of America, Inc.
595 Lawrence Expressway
Sunnyvale, CA 94085
Phone: 1-408-530-4537. Fax: 1-408-530-4515