Re: apache style squid?

From: Dean Gaudet <dgaudet-list-squid-dev@dont-contact.us>
Date: Tue, 7 Oct 1997 19:01:57 -0700 (PDT)

--MimeMultipartBoundary
Content-Type: TEXT/PLAIN; charset=US-ASCII

On 8 Oct 1997, Michael O'Reilly wrote:

> Oskar Pearson <oskar@is.co.za> writes:
> > Apache has a cache-module... perhaps it's worth spending time on,
> > rather than re-writing large amounts of squid?
>
> That's an interesting thought. I'll take a look at that. Wonder how it
> handles the cache index.

Uh, don't bother. It'd be better off if it was rewritten from scratch.
The Apache proxy module has numerous race conditions and hasn't ever been
tuned for performance, and probably isn't even correct for that matter. I
always recommend people use Squid for proxying.

> > I know that Dave Miller (one of the linux gurus) was very impressed with
> > the things squid does to try and optimise it's throughput. If you
> > compare something like the Zeus web server (http://www.zeus.co.uk/) with
> > apache there is a major speed difference... (300% supposedly).
> > Zeus is a single process which (from what I can see from the binary)
> > forks off as many processes as there are CPU's and then hands off
> > requests to these single, monolithic programs... if there is only one
> > CPU (and thus one process) and it starts running short on FD's, it seems
> > to fork off a seperate process to handle those requests.
>
> That's something close to the combined threads/process model someone
> else mentioned that looks like it would be a pretty neat idea.

The JAWS papers at <http://www.cs.wustl.edu/~jxh/research/research.html>
especially this one:
<http://www.cs.wustl.edu/~jxh/research/papers/infocom98-jaws.ps.gz>
go over some nice analysis of the various models.

-- 
Dean Gaudet, Performance Analyst, Transmeta Corp.
--MimeMultipartBoundary--
Received on Tue Jul 29 2003 - 13:15:43 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:11:26 MST