On Thu, Jul 05, 2007, Alexandre Correa wrote: > i have 1 dedicated server for squid serving about 700 users simultaneous and > 80 req/s !! Smallfry. :) > server is dual opteron dual core and 4gb of ram.. > > this cache_mem is fine ? > > cache_mem 256 MB > > or is best decrease ou increase this ? Thats fine. Just make sure you've either compiled Squid 64-bit or, if its compiled 32-bit, you don't exceed whatever the system-wide process max size is (2 or 3 gig, its system and kernel dependant.) The rest of your RAM will be used for kernel disk/socket buffers and such; that'll be good enough under your proposed load. You only need to up cache_mem if your workload thrashes the memory cache. Adrian
