[2934] in linux-net channel archive
Re: More on slow TCP in the pre-2.0 kernels
daemon@ATHENA.MIT.EDU (Alan Cox)
Fri May 17 17:04:51 1996
From: Alan Cox <alan@cymru.net>
To: schenk@rnode84.cs.toronto.edu (Eric Schenk)
Date: Fri, 17 May 1996 13:16:54 +0100 (BST)
Cc: linux-net@vger.rutgers.edu
In-Reply-To: <199605170643.CAA00914@rnode84.cs.toronto.edu> from "Eric Schenk" at May 17, 96 02:43:11 am
> In particular, now that we have fast retransmits and various corrections
> to the Jacobson slow start and congestion avoidance code, it turns out
> that if we are transmitting over highly congested links our performance
> goes down from 1.2.13 levels.
I'd expect that.
> congestion window between 1 and 2, and when fast retransmit gets triggered
> we will inject a long sequence of packets into the network which will all
> be lost. This just generally results in very slow behavior under these
What would happen if we didnt do fast retransmit with a congestion window
under 3 frames. At that point fast retransmit isnt much of a win anyway ?
Have I missed anything silly here ?
Alan
