nanog mailing list archives
Re: PC Bozo's World bites again (CNN, too)
From: Vern Paxson <vern () ee lbl gov>
Date: Fri, 29 May 1998 00:01:57 PDT
Has anyone done an analysis of how having a smaller MTU affects slow-start? It seems to me that a TCP session would come up to speed about 3 times faster with a mtu of ~500 vs a mtu of ~1500.
It doesn't, because the smaller MTU doesn't mean you get more acks per unit time, just that you get more acks overall. So you still get the same slow start increase pattern, only your segments are 1/3 the size they could be, so your data rate is 1/3 what it could otherwise be (until cwnd is big enough to consume the available bandwidth). The improved-performance-with-low-MTU is a puzzle I'd certainly like to understand better. It seems step 1 is to eliminate the possibility that the 1500 MTU is leading to fragmentation. Vern
Current thread:
- PC Bozo's World bites again (CNN, too) Vadim Antonov (May 27)
- Re: PC Bozo's World bites again (CNN, too) Karl Denninger (May 27)
- Re: PC Bozo's World bites again (CNN, too) Dean Robb (May 28)
- Re: PC Bozo's World bites again (CNN, too) Matthew Marlowe (May 27)
- Re: PC Bozo's World bites again (CNN, too) Michael Dillon (May 27)
- Re: PC Bozo's World bites again (CNN, too) ken emery (May 27)
- Re: PC Bozo's World bites again (CNN, too) Vadim Antonov (May 28)
- Re: PC Bozo's World bites again (CNN, too) Darrell Fuhriman (May 28)
- Re: PC Bozo's World bites again (CNN, too) Richard Irving (May 28)
- Re: PC Bozo's World bites again (CNN, too) Forrest W. Christian (May 28)
- Re: PC Bozo's World bites again (CNN, too) Michael Dillon (May 27)
- Re: PC Bozo's World bites again (CNN, too) Karl Denninger (May 27)
- <Possible follow-ups>
- Re: PC Bozo's World bites again (CNN, too) Vern Paxson (May 29)
- RE: PC Bozo's World bites again (CNN, too) Doug Stanfield (May 29)
- RE: PC Bozo's World bites again (CNN, too) Jeff Victor - Sun Microsystems Systems Engineer (May 29)