Re: [squid-users] squid + epoll polygraph test

I've been using squid3 with epoll support for a couple of months.
In my case, squid with poll/select did consume up to 100% CPU. With epoll, CPU usage dropped to less than 10%.

Long term average & max CPU usage:
http://webs.uolsinectis.com.ar/garana/x/cpu.4.png

With epoll, CPU usage over the last 24 hours:
http://webs.uolsinectis.com.ar/garana/x/cpu.png

Regards,

Gonzalo

On Tue, 2 Nov 2004 14:47:30 +0100 (CET)
Henrik Nordstrom <hno@squid-cache.org> wrote:

>
> On Tue, 2 Nov 2004, Muthukumar wrote:
>
> > Is there anyone benchmarked squid+epoll() on polygraph? How may I expect requests satisfaction limit on Linux host
> > 2.6.5-1.358 #1 i686 athlon i386 GNU/Linux platform?
>
> There has not been any benchmark on Squid-3 + epoll in a long time. The
> performance of this is not known.
>
> The Squid developers is currently focused on first getting Squid-3
> reasonably stable and correct before looking at performance.
>
> > During polygraph testing, I am getting errors as,
> > 004.03| ./Xaction.cc:79: error: 1/1 (267) unsupported HTTP status code
> > 004.03| ./Xaction.cc:79: error: 2/2 (267) unsupported HTTP status code
>
> Could be many things.
>
> I would recommend starting with a Squid-2.5 to verify that you have the
> Polygraph setup correct. This should run without any errors except the
> expected ones..
>
> Then try out Squid-3.
>
> Regards
> Henrik
>
Received on Tue Nov 02 2004 - 08:12:22 MST