[squid-users] Squid tuning

Hi,

I'm planning to change the hardware of our internet proxies (we're using
a two level proxy architecture) and benefits of this operation to also
change the OS and squid version ; new Linux (FC 3), new kernel (2.6.10)
and
concerning the hardware : Bi-Xeon 3 Ghz + 6*.36.5 Go Ultra 320 + 2 GB
RAM
+ gigabit network card. Each proxy having to handle about 80 req/s
during
peak times, and about 10000 users (up to 16000 according to the
cachemgr.cgi).

So I have a few questions concerning the new configuration :
- I plan to set up three RAID 1 logical disks, 1 for the OS, and
two for the cache dir. Do you think that's ok? What would be the
recommandation
for such a hardware configuration?
- How many cache dir do you recommend to use? If several, each one on a
hard drive?
- What file system should I use for the cache dir : ext2, ext3 or
reiserfs? With diskd, aufs?
- I were using lfuda till now, do you think I should change of policy
replacement?
- Is there a link beetween the size of the cache_dir and cache_mem
settings and
the size of the swap partition?

If anybody as any more "tips" or links on how to get the maximum
performance of
a squid box, I would be very interested.
Thanx in advance.

        Pierre-E
Received on Wed Mar 23 2005 - 02:45:49 MST