Re: [8023-CMSG] Fw: [802.1] Slides for per-priority flow control
Matt,
I didn't mean to imply there was poor cooperation between 802.1 and
802.3 in general. What I meant to imply was the problem space is bigger
than the territory covered by both 802.1 and 802.3 and any effective
solutions for the "limited" Ethernet interconnects we are trying to
enable will likely require support by both the 802.3 and 802.1
standards. I think the CMSG efforts have progressed to the point that we
are close to agreeing on where the split between functional
responsibilities should lie (which was necessary for getting us beyond
the major contention over turf). Whether or not the CMSG can make a good
enough case for tweaking one or both standards is still TBD. But we will
have to "think" outside the 802.3 (and 802.1) box to do so. You may want
to wait until we propose changes and show their affects through
simulation before declaring they break the architecture.
Gary
-----Original Message-----
From: owner-stds-802-3-cm@listserv.ieee.org
[mailto:owner-stds-802-3-cm@listserv.ieee.org] On Behalf Of Matt Squire
Sent: Wednesday, July 21, 2004 10:57 PM
To: STDS-802-3-CM@listserv.ieee.org
Subject: Re: [8023-CMSG] Fw: [802.1] Slides for per-priority flow
control
> It seems to
> me there was an expectation of a more cooperative relationship between
> 802.1 and 802.3.
>
In my experience, there has been a very cooperative relationship between
802.1 and 802.3. The two groups communicate often and fairly well.
What you may be experiencing, instead of a lack of cooperation between
802.1 and 802.3, is a non-trivial number of people who disagree with the
concepts you are proposing. One of the areas of disagreement (there
were several) with the proposals is that they violate the existing
architectures of 802.1 and 802.3, where the proposals suggestion 802.3
"look at" information that is generally in "802.1"'s domain.
So rather than lack of cooperation, an alternate way to view the
discussions last week is that 802.1 and 802.3 were in complete agreement
- both groups are providing feedback that the suggestions are
architecturally broken, no matter which group looks at them.
- Matt