[1253] in linux-net channel archive

home help back first fref pref prev next nref lref last post

Re: Asymmetric TCP/PPP performance (very poor receiving TCP)

daemon@ATHENA.MIT.EDU (Rob Janssen reading Linux mailingl)
Fri Oct 27 06:39:47 1995

From: linux@pe1chl.ampr.org (Rob Janssen reading Linux mailinglist)
To: rlk@tiac.net (Robert L Krawitz)
Date: Wed, 25 Oct 1995 23:16:36 +0100 (MET)
Cc: linux-net@vger.rutgers.edu
In-Reply-To: <199510241418.KAA06787@sunspot.tiac.net> from "Robert L Krawitz" at Oct 24, 95 10:18:44 am
Reply-To: linux-vger@wab-tis.rabobank.nl

According to Robert L Krawitz:
> tcpdump shows that when receiving data I send out a lot of ack's whose
> sole purpose seems to be to increase the window after ftp pulls each
> chunk of data off.  Given that the window's never falling very low,
> these ack's seem excessive and probably degrade bandwidth.  In these
> traces, rlk-ppp-1 is my machine and bloch is the remote machine.
> 
> I stared at the TCP code for a while but didn't make any headway into
> this.  Is this problem known?  Has anyone else seen it?

Yes, I have seen this.  Linux sends too many ACKs, it should send the
second ACK only when the window increases from 0 (or maybe from a very
small value) when data is read.
The extra flow of ACKs is also very noticible on the (low speed) amateur
packet radio network :-(

However, I'm not sure it is the cause of the problem you are seeing.

Rob

-- 
+------------------------------------+--------------------------------------+
| Rob Janssen         rob@knoware.nl | AMPRnet:   rob@pe1chl.ampr.org       |
| e-mail: pe1chl@wab-tis.rabobank.nl | AX.25 BBS: PE1CHL@PI8WNO.#UTR.NLD.EU |
+------------------------------------+--------------------------------------+

home help back first fref pref prev next nref lref last post