Re: [squid-users] Squid Efficiency - What else to tweak?

From: Matus UHLAR - fantomas <uhlar@dont-contact.us>
Date: Thu, 23 Feb 2006 10:50:36 +0100

On 21.02 11:57, Denis Vlasenko wrote:
> On Tuesday 21 February 2006 10:31, Matus UHLAR - fantomas wrote:
> > On 21.02 12:33, Ow Mun Heng wrote:
> > > I've got a squid cache running as transparent proxy on an _very_ old
> > > machine.
> > >
> > > P 133 Mhz
> > > 128MB Ram
> > > 20GB Hard Disk
> > > 1GB Cache in aufs
> > > Fedora Core 3
> > >
> > > Here are the stats reported by calamaris 2.99
> > > Total amount: requests 10471
> > > Total amount cached: requests 1330
> > > Request hit rate: % 12.70
> > > Total Bandwidth: Byte 107M
> > > Bandwidth savings: Byte 1767K
> > > Bandwidth savings in Percent (Byte hit rate): % 1.62
>
> I think he is mostly worried about having only 1.62% of traffic
> saved by squid by suppying cached content instead.
>
> What veterans are doing to get it to some decent numbers?

...add more memory and disk space. I haven't seen his cache_mem and memory
usage, but maybe using more than one GB for disk cache would help a bit
without machine starting to swap...

But as long as the question was what to do "besides the hardware"...

Also: What replacement policy do you (Ow Mun Heng) use? I use heap LFUDA for
disk objects and heap GSDF for memory cache. LFUDA gave me 2% better byte
radio and GSDF 2% better hit ratio IIRC.

-- 
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.
I feel like I'm diagonally parked in a parallel universe. 
Received on Thu Feb 23 2006 - 02:50:40 MST

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