OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
Subject: Re: performance issues
From: Brad Knowles (blkskynet.be)
Date: Fri May 19 2000 - 07:41:33 CDT

At 1:22 PM +0200 2000/5/19, Brad Knowles wrote:

> I'm going to try boosting the maximum size of the active
> queue, and increasing allowed parallelism, etc.... We'll see
> how high we can crank up these numbers.

        Hmm. None of that seemed to make a difference. I increased the
maximum size of the active queue to 2000, I increased both the local
and default concurrency to 2000, I increased the default maximum
number of processes to 500 (then 2000), I set the
qmgr_site_hoq_factor to 90, and none of it seemed to make any
signficant difference -- not even with syslogd turned off and the
postfix mqueue on tmpfs. I never even went over 75 total processes
on the system (I typically have about 22 when the machine is idle).

        No matter what I do, I can't seem to do any better than this:

                /usr/bin/time ./smtp-source -m1000 -s20 -l15360 -c
-tnulllocalhost localhost
                1000

                real 17.3
                user 0.9
                sys 2.4

        For about 57.803 messages per second, which only works out to
867.045 KBytes/second, and I know that this machine should be able to
do quite a bit more than that on a 100Base-TX network.

        I just don't get it. 

--
   These are my opinions -- not to be taken as official Skynet policy
======================================================================
Brad Knowles, <blkskynet.be>                || Belgacom Skynet SA/NV
Systems Architect, Mail/News/FTP/Proxy Admin || Rue Colonel Bourg, 124
Phone/Fax: +32-2-706.13.11/12.49             || B-1140 Brussels
http://www.skynet.be                         || Belgium