Re: [squid-users] Performance problems - need some advice

From: Matus UHLAR - fantomas <uhlar@dont-contact.us>
Date: Tue, 14 Feb 2006 23:17:24 +0100

On 08.02 09:13, Kinkie wrote:
> On Tue, 2006-02-07 at 16:29 -0800, Jeremy Utley wrote:
> > /dev/sdb1 -> /cache1
> > /dev/sdc1 -> /cache2
> > /dev/sdd1 -> /cache3
> > /dev/sde1 -> /cache4
> > /dev/sdf1 -> /cache5
> > Each one has it's own cache_dir line in the squid.conf file.
>
> You might want to double them: each cache_dir has its own server thread
> AFAIK. Your high iowait stats mean that the threads get blocked while
> waiting for i/o. Having more worker threads might mean higher
> parallelism and less iowait.

I doubt so. The high iowait means that the disks are overloaded, and putting
more cache_dirs won't make them loaded less, but even more (and will bring
more overhead in squid and OS too)

> So:
> cache_dir aufs /cache1/a <blah>
> cache_dir aufs /cache1/b <blah>
> cache_dir aufs /cache2/a <blah>
> etc etc etc.

> Squid can help with the hot object cache in RAM, but that cache can
> cover maybe 0.5% of your total served content, which is not much
> anyways. Paradoxically it might help you to DECREASE your cache_dir
> sizes, in order to make it so that squid keeps warmer data on-disk (and
> in RAM) and goes to the backend server to fetch less-popular contents.
> In other words, to decrease your cached objects lifetime.

However, if you have enough of RAM, using bigger cache_mem with low
maximum_object_size_in_memory might help - I have 150MB of cache_mem of 128kB
objects max and have wuite high memory hit ratio.

-- 
Matus UHLAR - fantomas, uhlar@fantomas.sk ; http://www.fantomas.sk/
Warning: I wish NOT to receive e-mail advertising to this address.
Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu.
Your mouse has moved. Windows NT will now restart for changes to take
to take effect. [OK]
Received on Tue Feb 14 2006 - 15:17:34 MST

This archive was generated by hypermail pre-2.1.9 : Wed Mar 01 2006 - 12:00:03 MST