What about RAID 0? Has anyone done in comparison vs multiple cache dirs?
> > I'm currently running Squid-2.4.stable2 on a dual 866mz P3, 512mb RAM,
> > scsi raid cache dir, Redhat 7.1 (2.4.2-2smp) . I'm seeing squid run on
> > only a single processor while I have about 200+ users hitting it. Is
> > there anything that I might be doing wrong? I'm using the following
> > during configuration of the squid source:
>
> Squid is a single-threaded process. It will only occupy one processor. It
> might jump across cpus but that is a function of the scheduler and any
> decent scheduler will try to leave a process on the same cpu to avoid
> invalidating processor cache memory. Using RAID is not the best way to
> manage squid. You are much better off spreading multiple cache_dirs across
> disks, one cache_dir per disk. Squid will spread the load for you.
>
> Colin
> --
> Colin Campbell
> Unix Support/Postmaster/Hostmaster
> CITEC
> +61 7 3006 4710
>
Received on Tue Sep 25 2001 - 18:55:44 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:02:29 MST