httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Vivek Sadananda Pai <vi...@cs.rice.edu>
Subject Re: [PATCH] 10x performance increase patch #9
Date Wed, 03 Nov 1999 22:59:58 GMT
> The base for this patch is Apache/1.3.6 + 10x performance increase
> patches #1 through #8.  After I contribute the final patch I'll port
> all my patches to the latest Apache release.

I've been sporadically following this discussion, but I have to admit
that I'm having a hard time putting it in context. My interests are
also more on absolute performance rather than flexibility, so I'm
probably not trying to achieve the same thing as the Apache
developers.

I've been trying to figure out what yields this 10x improvement, and
how the final result fares in absolute terms. The latest patch said:

> +Here is the configuration I use to achieve maximum SPECweb96 
> +performance on an SGI Origin200 server running Irix 6.5 with two 
> +processors, two gigabytes of memory, and four 100BaseT network 
> +interfaces. Without any patching or tuning Apache on this system 
> +handles 240 operations per second; with my patches and the following 
> +tuning it handles 2400*.

Is this configuration essentially the same as

http://www.spec.org/osg/web96/results/res98q3/web96-980729-03005.html

There, the system got 2753 reqs/sec using Netscape FastTrack 3.02.
Incidentally, those were the most recent SpecWeb99 results I found
from SGI - available at http://www.spec.org/osg/web96/results/res98q3/

So, it appears that the absolute performance is good for Apache, but
nothing earth-shattering. That's not overly surprising. So, that means
that whatever was 10x slower was pretty slow. Was it just a completely
untuned Apache? What were the performance advantages of the new code
versus the advantages gained just from tuning?

-Vivek


Mime
View raw message