Re: Linux and Apache performance, Update

H. Peter Anvin (hpa@transmeta.com)
4 Jun 1999 21:39:12 GMT

Followup to: <E10pvHp-00031s-00@the-village.bc.nu>
By author: Alan Cox <alan@lxorguk.ukuu.org.uk>
In newsgroup: linux.dev.kernel
>
> > So the common, valid, comparison most real people will want to see is
> ^^^^^
>
> Unfortunately they aren't valid because Apache is designed for the real
> world not test labs. The test lab proceedures used by all parties I know of
> are an absolute joke and people now are forced by marketing to write
> web servers that suck for real use but do the benchmark well
>
> What kills real world http performance is _slow_ connections, fast ones are
> actually less of an issue.
>
> Linux can do very high speeds with Zeus and with Zab's new httpd, but as
> I suggested to Zab, his httpd ought to come with an "optimised for benchmarking"
> sticker.
>
> IMNSHO there are no valid comparison benches available. That is a problem and
> not just for Apache.
>

One interesting thing is that Network Applicance's tanks can speak
HTTP for static data serving. They then recommend a tranditional host
for the dynamic content. There should be no reason why you couldn't
run a super-fast static page server *and* a good dynamic server at the
same time.

-hpa

-- 
"The user's computer downloads the ActiveX code and simulates a 'Blue
Screen' crash, a generally benign event most users are familiar with
and that would not necessarily arouse suspicions."
-- Security exploit description on http://www.zks.net/p3/how.asp


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/