Benchmarking Terminology for Resource Reservation Capable Routers
draft-ietf-bmwg-benchres-term-08

The following can be fixed by the RFC editor:

(Comments by Al Morton )

I think there's clear improvement.  I noticed that the names of
states in section 3.5.5 has changed from lossy and loss-free to
"good" and "bad".  We strive to avoid using terms associated with
value judgements, like pass and fail.  But this is easily solved
by substituting a different pair of terms.

[Ballot discuss]
1. It is not clear if the document really intends to be restricted to IP routers supporting IntServ as the abstract says. Would for example NSIS qualify as a Resource Reservation Protocol as defined in Section 3.2? I would prefer either to eliminated the text restricting the applicability to IntServ, or to clarify (title included) that this document is for IntServ devices only.

2. Section 3.4.1, 3.4.2 and 3.4.3 define 'seconds' as units for latency. This does not seem to be granular enough and thus conflicts with Section 3.8 in RFC 1224 which requires that time units for latency be 'Time with fine enough units to distinguish between two events'.

3. All Section 3.5 seems to have been written with an orientation that better fits routers in general and not QoS routers. Loss-free, lossy conditions and the events that can trigger the transition between one and the other are not the only or the most appropriate metrics to characterize a QoS router. In fact when looking at what defines scalability, the capability to sustain application flows within the appropriate QoS performance parameters is more important than eventual loss in applications like VoIP for example. I suggest that this section modifies and adds appropriate text to reflect these conditions.

From one of the WG chairs:

-----Original Message-----
From: Al Morton [mailto:acmorton@att.com]
Sent: vrijdag 2 januari 2004 18:11
To: Wijnen, Bert (Bert); Kevin Dubray (E-mail)
Subject: Re: Status -- benchres


At 04:06 PM 01/02/2004 +0100, Wijnen, Bert (Bert) wrote:
>And finally we have "AD is watching, revised ID needed"
>again by Randy back in August:
>  draft-ietf-bmwg-benchres-term-03.txt

benchres-term-03 is back in the WG's court.
There were substantial revisions in this version,
but I thought some of the comments could have been implemented
more thoroughly, and gave that feedback.  We're
waiting for 04 to do a WGLC. Zero comment on the list.
(part of IETF-57 meeting minutes below)

Al

2nd Session
Tuesday, July 15, 2003, 1415-1515
=======================================

7.  Resource Reservation Terminology benchmark I-Ds --  (Fehér Gábor Gume)

http//www.ietf.cnri.reston.va.us/internet-drafts/draft-ietf-bmwg-benchres-term-03.txt

Description of changes based on AD review, primarily to remove Diffserv
material, sync-up with current NSIS drafts, and remove non-standard examples
(e.g., Boomerang). There were still a few places where comments could be more
thoroughly adopted, and the draft will be revised again shortly.
Gábor made an appeal for more readership and review on BMWG-list.

Date: Wed, 05 Nov 2003 16:54:14 -0500
To: Randy Bush , Kevin Dubray
From: Al Morton
Subject: Re: BMWG I-D in state Expert Review
Cc: Bert Wijnen


I think benchres-term-03 is back in the WG's court.
There were substantial revisions in this version,
but I thought some of the comments could have been implemented
more thoroughly, and gave that feedback.  I think we're
waiting for 04 to do a WGLC...

Date: Wed, 19 Feb 2003 17:58:54 -0500
From: Kevin Dubray
To: Randy Bush
CC: Bert Wijnen
Subject: Re: draft-ietf-bmwg-benchres-term-02.txt

I need to contact the author and see what we can do. (E.g., excise
subjective observations regarding DiffServ; relegate non-RSVP
elements to an appendix; see what can be done to reconcile the
I-D to the NSIS works in progress; tighten focus.)