Re: [squid-users] optimizing squid and FreeBSD

tis 2007-03-20 klockan 21:09 -0300 skrev Michel Santos:

> I do compare the incoming http traffic to the outgoing. Higher the
> difference better my cache performance right.

The better hit ratio you have. But tells nothing about the performance.
An overloaded disk can be significantly slower than a fast Internet
connection.

> that is certainly an interesting point. IO Bound I guess can be fight by
> faster and cpu independent disks and subsystem (scsi) and then using
> polling on for example em (intel pro) nics which seem to "produce" less
> interrupts.

None of these helps in speeding up the rotation and seek time of a
disk.. It's physical limitaions of things moving around.

> Also setting vfs.write_behind and vfs.vmiodirenable may give important
> improvement on some hardware together with vfs.read_max.

Not familiar with FreeBSD terminology.

> All this does not cut ufs's bottleneck but helps a lot. So sure diskd is
> the preferred cache_dir on FreeBSD.

I would say aufs is the preferred cache_dir on FreeBSD, Linux and
Solaris these days.

aufs requires POSIX kernel threads, which is available even on FreeBSD
these days.

> But again, not on low traffic machines
> where I can not find any difference. IMO so long as your machine does not
> handle more than 2mb/s it does not matter what you do FreeBSD does it
> well either way - supposed you have good hardware.

With only 2mb/s you are unlikely to reach even 30 req/s, so yes.. and
this not even needing good hardware just not too crappy hardware.

Regards
Henrik