Re: Delay-pools limit more than specified

From: Borek Lupomesky <Borek.Lupomesky@dont-contact.us>
Date: Thu, 18 Mar 1999 09:31:49 +0100 (CET)

On Thu, 18 Mar 1999, David Luyer wrote:

> > This should impose limit of 8 KB/s aggregate bandwidth for "clients".
> > But in fact it seems that it limits it to half the value, ie. 4 KB/s (on
> > large downloads). Have I missed something? I tested this on a "test"
> > installation with single client.
>
> Do you actually have 64kbps bandwidth to the source site in question, and
> can all the machines in question handle it? (Eg, LC-III Macs max out at
> around 40kbps, as can NFS writes to Sun3 servers at even lower figures).

> (I'm not sure what could be causing this, and it could actually be a bug,
> but I'd like to know it wasn't some other thing causing it first.)

   It's a local "test bed" configuration running over 100baseTX. All
client, proxy and the http server are on local Ethernet. All machines are
Linux. Maybe I should do some more rigid measuring...? Also I tried other
bandwith allowances and the throughput has always converged to half of
what I specified.

   Bye Borek

--
=====================================================================
BOREK LUPOMESKY, network administrator    University of J. E. Purkyne
WWW:       http://www.ujep.cz/~lupomesk/  Ceske mladeze 8
IRCnet:    Borek @ #usti                  Usti nad Labem, 40096
talk:      borek@ishtar.ujep.cz           The Czech Republic
PGP keyid: 9DD1C54D   ICQ: 10139578       tel: +420-602-376368
==========[ MIME/ISO-8859-2 & PGP encrypted mail welcome ]===========
Received on Thu Mar 18 1999 - 01:53:32 MST

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:45:19 MST