On Aug 9, 2009, at 11:53 PM, Steven Blake wrote:
On Sat, 2009-08-08 at 00:29 -0700, Lars Eggert wrote:
Hi,
these are all good points. I wanted to repeat something I said at the
mike in Stockholm which I think is essential:
We cannot make a recommendation for a certain buffer size without
also
saying which queuing scheme we recommend for that buffer.
If we assume that FIFO queuing will be used, yes, the buffer size has
a large, direct impact on the observable behavior. Not so for various
AQM schemes (RED, etc.). Since my personal preference would be to
recommend something smarter than FIFO, I don't see buffer size as
very
critical then. (Now, how to correctly parameterize an AQM scheme
for a
given link, *that* would be critical...)
802.1Qbb (http://www.ieee802.org/1/pages/802.1bb.html) priority PAUSE
has the right semantics to push the queues back into the hosts where
they can react intelligently. Unfortunately, this would require
more-or-less universal adoption in modems, gateways, switches, hosts,
and WiFi APs.
Well, if you can put the algorithms into the home gateway to know when
to send PAUSE, you can by definition manage the queues intelligently
in the home gateway itself. I don't see how this changes anything -
unlike in localized L2 networks, in an IP network bottlenecks can be
anywhere on the path. There's nothing special about the second hop
that pushing the congestion notification one hop upstream solves
anything.
Take this to the extreme (no pun intended) and you wind up with X.25
backward congestion notification.
Cheers, DaveO.
Regards,
/////////////////////////////////////////////
Steven Blake [email protected]
Extreme Networks +1 919-884-3211
_______________________________________________
homegate mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homegate
