Re: SMP scalability goal

On Thu, 2008-04-24 at 16:09 -0300, Gonzalo Arana wrote:
> On Thu, Apr 24, 2008 at 1:31 PM, Alex Rousskov
> <rousskov@measurement-factory.com> wrote:
> >
> > On Thu, 2008-04-24 at 10:27 -0600, Duane Wessels wrote:
> > >
> > > On Thu, 24 Apr 2008, Alex Rousskov wrote:
> > >
> > > > I do not think it is realistic to expect nearly linear scale (close to
> > > > 100% and 200% increase in performance), especially for the first
> > > > implementation.
> > >
> > > As you know, disk is usually the bottleneck. So I think your goals
> > > and expectations should state whether disk caching is involved or
> > > not. Assuming disk caching is involved, maybe even state what
> > > storage type and a typical hit ratio.
> >
> > Very good point. I should have said that for this particular question
> > let's consider the "pure" case of no caching (neither disk nor memory).
>
> I believe another thing to consider is what if sysadmin configure
> squid with a very long url_regex acl (and use it in, say,
> http_access).
>
> Perhaps heavy acls should also be excluded from the metric.

I am interested in a simple, bare bones setup and ballpark estimates at
this point. There are lots of things that can be added on top of that,
of course, and some of those things can be optimized to benefit from SMP
support, but I am looking for a baseline estimate and relevant
performance data from other servers.

Thank you,

Alex.
Received on Thu Apr 24 2008 - 20:23:38 MDT