Thread Links | Date Links | ||||
---|---|---|---|---|---|
Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
One is not supposed to reorder packets of the sameThe above is covered in 802.1D section 7.7.3.
priority with the same DA/SA combination in an 802
network. I think this is specified in an 802.1 document
somewhere I will try to look it up to be sure.
-Sushil
Mike Takefman wrote:
Guys,sorry to jump in in mid debate and I admit I have not read
all of the answers in detail but....One is not supposed to reorder packets of the same
priority with the same DA/SA combination in an 802
network. I think this is specified in an 802.1 document
somewhere I will try to look it up to be sure.mike
Anoop Ghanwani wrote:
>
> Spencer,
>
> We keep getting back to a single TCP connection. This is not
> one connection. This is many connections bundled together.
> We don't have control of which individual
> connections get reordered. For all we know, the connection
> with the least traffic will get affected.
>
> In any case, this discussion is probably not appropriate for
> for 802.17...debating ordered delivery may be more
> appropriate for something like the end2end-interest list where
> we have folks that are experts on this issue. They understand
> the issues and can explain the reason for recommending
> in-sequence delivery much better than I can.
> Their work tells us to minimize, if not eliminate, out-of-order delivery.
> It is better to drop packets than to routinely deliver them
> out-of-order (as opposed to delivering out-of-order when some
> failure or topology change occurs, which is a rare event).
> TCP flows are by nature greedy so you will
> routinely get to the point where you're out-of-profile.
>
> -Anoop
>
> > -----Original Message-----
> > From: Dawkins, Spencer [mailto:Spencer.DAWKINS@xxxxxxxxxxxxxxx]
> > Sent: Friday, December 07, 2001 8:14 AM
> > To: stds-802-17@xxxxxxxx
> > Subject: RE: [RPRWG] Frame reordering for medium priority
> > traffic in Ganda lf?
> >
> >
> >
> > Sorry, but I'm still confused. How can reordering
> > out-of-profile TCP traffic make things worse? Either it
> > triggers fast retransmit/fast recovery, so the sending
> > station slows down its sending rate, or it doesn't trigger
> > fast retransmit/fast recovery, so there is no effect.
> >
> > Spencer
> >
> > -----Original Message-----
> > From: Anoop Ghanwani [mailto:anoop@xxxxxxxxxxxxxx]
> > Sent: Friday, December 07, 2001 10:05 AM
> > To: 'Dawkins, Spencer'; stds-802-17@xxxxxxxx
> > Subject: RE: [RPRWG] Frame reordering for medium priority
> > traffic in Ganda lf?
> >
> >
> >
> > First, this is not a single station that we're talking about here.
> > Second, the out-of-profile traffic was admitted because the
> > network thought it could deliver it. Reordering the traffic might
> > actually make things worse. It's like getting a freebie that might
> > make your life worse without telling you about it.
> >
> > -Anoop
> >--
Michael Takefman tak@xxxxxxxxx
Manager of Engineering, Cisco Systems
Chair IEEE 802.17 Stds WG
2000 Innovation Dr, Ottawa, Canada, K2K 3E8
voice: 613-254-3399 fax: 613-254-4867
begin:vcard n:Pandhi;Sushil tel;work:845-731-2256 x-mozilla-html:FALSE org:Fujitsu FNC;Systems Engineering adr:;;2 Blue Hill Plaza;Pearl River;New York;10965;USA version:2.1 email;internet:sushil.pandhi@xxxxxxxxxxxxxxx title:Principal Systems Engineer fn:Sushil N. Pandhi end:vcard