httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chuck Murcko <>
Subject Re: roy's l_c perf patch and spareservers
Date Sat, 15 Feb 1997 05:15:13 GMT
Marc Slemko wrote:
> On Fri, 14 Feb 1997, Dean Gaudet wrote:
> >
> > The other thing I found, just like Chuck said a while back, I played with
> > the Min/MaxSpareServers settings and removed all the huge spikes from my
> > load averages.  The change in load *could* be related to locality of
> > reference in the L2 cache, but I'm not convinced that is the only thing.
> > We're talking a difference of 50% vs 20% "cpu usage" as reported by the
> > status module.  (I'd really like to revamp status to do 5 minute rolling
> > averages.)
> Played in what way?  Just increased them both?

Sort of. Start with Min/Max set to a factor of 2 difference, say, 4/8.
Double 'em until the CPU percentage indicated in the status display
drops no further (or only a little), and then leave 'em there. The
original ratio of 1/2 can be tuned up for servers under higher load, or
down for those under lower loads. Dropping Min too low results in
excessive forking and lousy response on load spikes; making Max too high
results in wasted machine resources. It's just solving a limit problem
iteratively. An initial ratio of 1/2 worked best for my moderate to high

You have to let the machine run a while, how long depends on time of day
if the load varies in the usual sinusoidal way over each 24 hr. period,
and how well you can extrapolate. 8^)
Chuck Murcko
The Topsail Group, West Chester PA USA

View raw message