RE: [squid-users] Sizing a pretty Large Squid Install

From: Chemolli Francesco (USI) <ChemolliF@dont-contact.us>
Date: Thu, 10 May 2001 19:21:47 +0200

> Sorry, but no. You might see 70-80 from the first IDE 7200
> RPM drive.
> 30 from the second. 15 from the third...and so on. 10k RPM
> SCSI drives
> raise this by a small amount, 15k drives by another small
> amount. Squid
> does not scale linearly with number of drives. Many, many benchmarks
> will bear this assertion out...there is /no/ point arguing about it.
> Obviously the other hardware in the system can help squeeze
> more from a
> number of drives (more RAM is very helpful--which is why I always
> configure RAM very generously in our systems).

I've had a peak throughput of 197 hits/sec (average over an 1-hour period,
so the actual peak is higher). Hardware specs:
dual PIII/450
1 Gb RAM
6 9.1Gb (not sure about RPM, might even be 7.2k) Ultra2-wide spindles,
        1 for system, 5 for diskd-based (10 cache_dirs) cache on ext2
AMI MegaRaid controller used as JBOD.
2x eepro100 NICs with Intel drivers on Nortel switches

SW configuration:
Linux-2.2.19
Squid-2.4DEVEL3-NTLM
512Mb mem_cache

Only concession, a significant percentage of the requests are for
NTLM authentication, thus serviced very fast. On the downside, I
have a few 1500+-users ACLs and a fairly complex authorization
scheme (not extreme, but complex).

BTW: Hans Reiser has been bashing squid on the Linux kernel mailing list
today. He has some points, but he is blunt as he is known to be in
expressing himself.

-- 
	/kinkie
Received on Thu May 10 2001 - 17:02:52 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:59:56 MST