Re: [squid-users] Squid - performance

On 16.08 21:25, lokesh.khanna@accelonafrica.com wrote:
> I am using 2 Scsi hard disk of 36 GB for cache. I have assigned only 15
> GB ( less than 50 % of total disk ) for cache in each disk .
> So total cache is 30 GB.
>
> cache_dir diskd /cache1/squid 15360 16 256 Q1=64 Q2=72
> cache_dir diskd /cache2/squid 15360 16 256 Q1=64 Q2=72
>
>
> I want to know if squid uses full 30 GB for cache then what will happen?
> Shouldn't squid start deleting those cached objects which are not in
> use?
> Which parameter define that?

cache_swap_low and cache_swap_high define, when will squid start removing
objects and how much will it remove. cache_replacement_policy defines, how
will squid choose which objects to remove.

Note that for 36GB disks I have no problems defining cache_dir 30000
(the FS becomes filled up to ~86% which is OK)

> What parameters do I need to check to get best performance and caching.

increase cache size for bigger efficiency, but not too big (filesystems
filled up too much use to have low performance).

> My squid is currently handling 80 req/sec and during peak hours nearly
> 1500 users uses caching server.
> I can save maximum 15 % to 18 % bandwidth. Is it possible to save more
> bandwidth with better performance??

yes, do the above, choose good replacement policy (heap LFUDA or heap GSDF)
and increase maximum_object_size (for 2x36 GB I'd use 32 MB and maybe more).

-- 
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.
If Barbie is so popular, why do you have to buy her friends?