Date: Thu, 15 Aug 2002 01:33:21 -0400 From: Barney Wolff <barney@tp.databus.com> To: Mike Silbersack <silby@silby.com> Cc: Barney Wolff <barney@tp.databus.com>, Oleg Polyakov <opolyakov@yahoo.com>, freebsd-net@FreeBSD.ORG Subject: Re: Initial congestion window increase Message-ID: <20020815053321.GA37994@tp.databus.com> In-Reply-To: <20020814233935.F97690-100000@patrocles.silby.com> References: <20020814121701.GA27934@tp.databus.com> <20020814233935.F97690-100000@patrocles.silby.com>
next in thread | previous in thread | raw e-mail | index | archive | help
RFC2414-5 say no more than 4380, so your first example is wrong. And why do you assume that the jumbo stops at your border? Lots of people connect over non-local areas at gigE, these days. I don't know that they use jumbo, but I don't know that they don't. Internet congestion is measured in bits, not packets. Links don't care about packets and these days neither do routers. Generally, we should just do what the RFCs say, rather than trying to rethink every issue. It's really quite rare that the TCP RFCs are shown to be wrong, and the people who do it get famous. The OS's that deliberately flout the RFCs get famous too, in another way. On Thu, Aug 15, 2002 at 12:05:35AM -0500, Mike Silbersack wrote: > > Hrm, I'm not sure that PMTUD is a strong enough argument against X*MSS > slowstart for gigabit networks. Think about the following cases: > > 1. Server with MTU 1500, client with MTU 1480 (they're going over PPPoE > or something similar.) > > - All four 1500 byte packets sent back to back, all 4 bounced with ICMP > too big messages. Bandwidth wasted: All 4 packets traversing the net, > all 4 icmps coming back across the net. > > 2. Server with MTU 9000, client with MTU 1500. > > - All four 9000 byte packets sent back to back, bounced back at local > border router with MTU of 1500. Bandwidth wasted: Internal network > bandwidth only. Perhaps less than 4 packets, if all data fit into a > single 9000 byte packet. > > Considering this, I don't believe that the gigabit host using jumbo frames > would be any more harmful than a 100mbps host using normal ethernet > frames. -- Barney Wolff I'm available by contract or FT: http://www.databus.com/bwresume.pdf To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020815053321.GA37994>