Re: ACKing every packet [was RE: comments on draft-ietf-tcpsat-stand- mech-04.txt]
Vern Paxson <vern@ee.lbl.gov> Sat, 20 June 1998 06:21 UTC
Message-Id: <199806200621.XAA03739@daffy.ee.lbl.gov>
To: "Falk, Aaron" <Aaron.Falk@trw.com>
Cc: tcp-over-satellite@achtung.sp.trw.com
Subject: Re: ACKing every packet [was RE: comments on draft-ietf-tcpsat-stand- mech-04.txt]
In-Reply-To: Your message of Thu, 18 Jun 1998 11:29:13 PDT.
Date: Fri, 19 Jun 1998 23:21:20 -0700
From: Vern Paxson <vern@ee.lbl.gov>
Sender: owner-tcp-over-satellite@achtung.sp.trw.com
Precedence: bulk
Status: RO
Content-Length: 734
Lines: 14
> We all know that congestion control isn't fair to long delay > connections. It takes longer for the window to open to use available > bandwidth. Contrarily, when there is congestion, it takes longer for the > window to close--making long delay connections fairly unresponsive to > network congestion, i.e. the window is closed when it should open and > continues to open when it should close. However, the time constants themselves in the two cases are *way* different. The time it takes for the long delay connection to open its window is many RTTs; to close it, a single RTT. So yes, it responds to congestion in 500 msec instead of 120 msec - but as soon as it does, it won't be bothering you for a long time to come. Vern
- ACKing every packet [was RE: comments on draft-ie… Falk, Aaron
- Re: ACKing every packet [was RE: comments on draf… Eric Travis
- Re: ACKing every packet [was RE: comments on draf… Vern Paxson