Re: delayed ACKs for retransmitted packets: ouch! (fwd)

Neal Cardwell (cardwell@cs.washington.edu)
Wed, 4 Nov 1998 18:58:33 -0800 (PST)

---------- Forwarded message ----------
Date: Wed, 04 Nov 1998 15:58:49 PST
From: Vern Paxson <vern@ee.lbl.gov>
To: Neal Cardwell <cardwell@cs.washington.edu>
Cc: mallman@lerc.nasa.gov
Subject: Re: delayed ACKs for retransmitted packets: ouch!

> Ah, good idea. I added a trace from the receiver side. This trace is
> definitely more convincing.

Indeed - pretty stunning, what pessimal acking behavior Linux manages
to achieve in this case!

> Do you think it would be worth tweaking the congestion control I-D too?

Yes - I think that's better, actually (less work and a stronger statement).
Mark, what do you think?

Vern