Re: multiple squids

It was not my intention to patronise you, only that this approach has
been discussed before to quite some extent.

I am quite certain that most of what is to be said has not been said.

/Henrik

Chemolli Francesco (USI) wrote:
>
> > > Can't we just expand on the diskd concept?
> > > Idea: the parent process forks one diskd for each storage, and
> > > passes shmem IDs to all children. Then accesses are done
> > > through this helper processes farm.
> >
> > Look back in the squid-dev archives. Something similar to
> > this has been
> > proposed before when discussing multithreading, multiprocesses, and
> > other approaches to getting Squid running in SMP environments.
> >
> > Note that simply doing a diskd won't suffice as it is nothing
> > more than
> > simply asyncronous i/o with some minor mapping. Far more has to be
> > restructured for it to work.
>
> Which is exactly what I said below.
> Not that I assume to be particularly original :-)
>
> I won't expand on the subject further, assuming that all that there
> was to say has already been said. And being I not in the
> position to back my facts with code, I'll just leave it up
> to those who are.
>
> --
> /kinkie
Received on Tue Oct 17 2000 - 09:11:58 MDT