On Mon, Mar 24, 2008, Michael Gale wrote: > Hey, > > We are working on our hardware requirements and am looking for some > feedback. Please let me know what you think: > > Demand: > - 225 requests per second during peak times in 2008. So we are plaining > for 300 RPS minimal per server. Ideally if each server could handle 600 > RPS that would be good.
Not too bad on current hardware. Factor in about 80-100 req/sec per disk for a normal forward-proxy load. > - We have 1600 remote locations connected via sat link, each with about > 4 devices behind it. ok. > - 125GB per month of HTTP traffic > > We currently are planing on two servers being available behind an LVS > router. These two servers will speak with a squid instance at each > location so some form of peering can be used. ok. Just be careful how you distribute requests - you need to keep things like constant source address in mind when doing stuff or some things might subtly break. > So I have the following questions: > > 1. Would there be any problem with squid running at each sat location > (1600) trying to use a peering method with squidpeer.domain.com IP that > is load balanced by an LVS router pointing to two squid servers ? > > 2. Does squid benefit from a dual core or quad core setup at all ? Dual core, yes. Quad core, not so much. > 3. How do these hardware requirements look, per server: > - 4 drives for squid cache, hardware raid stripped > - 4ms seek time, 73GB of space =~ 294GB of cache available Don't raid them. Mirror, sure. Don't stripe or RAID5 them. > - Looking to use at least 150GB of cache per server ok. > - 8GB of RAM plenty. > - Two dual core or two quad core 3.0Ghz processors. Dual-core is enough. Squid can't take effective advantage of >1 CPU at the moment and your OS will use the other thread for network/disk IO. ADrian -- - Xenion - http://www.xenion.com.au/ - VPS Hosting - Commercial Squid Support - - $25/pm entry-level VPSes w/ capped bandwidth charges available in WA -
