[PWE3] PW performance measurements using VCCV

Yaakov, Tom and all,

I've read the just posted draft-nadeau-pwe3-perf-timing-measure-00.txt.
and I see that, among other things, you have proposed measurement of
PW packet loss rate (PLR) using a new format of VCCV packets (to be
defined).

In this regard I would appreciate your opinion on relevance of the problems 
raised in the (now expired but still available)
draft-rosen-pwe3-congestion-01.txt 
regarding ability to measure packet loss using "modified VCCV packets"
(Section 2)
to your current proposal.

Of course, that draft has dealt with a different problem: PW congestion
detection and not PW PLR measurement.

On the other hand, these two problems look to me reasonably similar; and one
may be more or less dificult than the other depending on the desired degree
of
accuracy of the PLR measurements and time intervals for which it is
measured.

IMHO other proposals in the draft (measurement of the PW packet delay and
PW packet delay variation using VCCV) also require specification of the
desired degree(s) of accuracy. E.g., I do not see how to select the most
appropriate timestamping format if the desired accuracy of these 
measurements is not specified.

Hopefully these notes will be useful.

Regards,
                          Sasha

_______________________________________________
pwe3 mailing list
pwe3@ietf.org
https://www1.ietf.org/mailman/listinfo/pwe3