On Wed, 26 Apr 2000, Leslie Mikesell wrote:
> According to Perrin Harkins:
> > On Tue, 25 Apr 2000 [EMAIL PROTECTED] wrote:
> > > With mod_proxy you really only need a few mod_perl processes because
> > > no longer is the mod_perl ("heavy") apache process i/o bound. It's
> > > now CPU bound. (or should be under heavy load)
> >
> > I think for most of us this is usually not the case, since most web apps
> > involve using some kind of external data source like a database or search
> > engine. They spend most of their time waiting on that resource rather
> > than using the CPU.
>
> If you have tried it and it didn't work for you, please post the
> details to help us understand your real bottleneck.
Sorry, I guess it was unclear what I was talking about. The proxy model
does work for me, but my mod_perl processes are not CPU bound because they
are still waiting for external data sources. I suspect most people are in
a similar situation.
> >The sequential model is great if you're the next in line,
> > but terrible if there are 50 big requests in front of you and yours is
> > very small. Parallelism evens things out.
>
> Or it just adds more overhead. If you have enough parallelism to
> keep your bottleneck busy, the 50th request can only come out slower
> by switching among jobs more often.
Everything gets slowed down some by context switching, but my point is
that a long request doesn't hold up short requests in the same way if you
are processing in parallel. Also, if all of your requests are n seconds
long, and you process 10 of them that arrive at the server simultaneously,
the parallel model makes the last request finish about the same time as
the first, as opposed to finishing 9 * n later than the first.
- Perrin
