Linux Benchmarking

[Fraser Campbell]

On Thu, 2005-18-08 at 00:29 -0400, Jason Carson wrote:

> I am thinking about doing a Linux comparison by benchmarking various
> distros then posting the results on my website.
> 
> Does anyone have any recommendations as to what software I can use to do
> the benchmarking. I found this page with a bunch of tests
> (http://lbs.sourceforge.net/)

If you want to do some benchmarking you might instead consider
benchmarking given applications (apache, postgres, postfix, ???) on your
distro of choice.  IMO benchmarking server apps is useful/interesting,
typical desktop apps just sit waiting for input, desktop "performance"
is not easy to measure.

Once you have baseline numbers for your server app of choice then try to
improve the results, performance tuning can be a lot of fun and you will
learn a lot.  Even though your findings are likely specific to your
hardware, software and workload they can still be of interest to others
(depending on what you benchmark).

Perf tuning is quite hardware specific.  For example, a given I/O
scheduler can be the beez knees on local SCSI/RAID disks but might be an
awful choice if you disks are SAN.

Perf tuning is also software specific, kernel 2.6.5 and 2.6.12 might
exhibit quite different performance characteristics and reactions to the
same tuning options.

-- 
Fraser Campbell <fraser-Txk5XLRqZ6CsTnJN9+BGXg at public.gmane.org>               http://www.wehave.net/
Georgetown, Ontario, Canada                       Debian GNU/Linux


--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml