First, a clarification would be needed. Are Squid and Apache running on each box? What is the hardware configuration on each server? What size are the files being served? Are the webpages dynamic or static? Is MySQL involved? Is squid currently deployed only as a proxy? or a cache? This is only a very small part of the equation to determine how much more performance you can squeeze out of your servers. Your question is begging for serious hands-on systems analysis. You will get a lot of answers from people on the list; but you need to have someone actually sit down and work with you one-on-one. On Wed, 1 Dec 2004 15:28:57 -0700, sweetgrass@zen2now.com wrote: > currently we are running an HTTP and HTTPS cluster of 6 RH ES 3 Dell 2650's lb > balanced with a coyote point E350. > > the scenario is, HTTP request comes into the LB which forwards it to one of > the 6 servers, squid pickups the request on port 80 forwards it to apache on > port 81 and apache answers the request, gives it back to squid and squid > sends it on its way via coyote point. this cluster currently receives around > 4mil hits a day and typically loads rise above 5 sometimes as high 30 on some > of the boxes. > > how could performance be tweaked, if at all? squid, httpd, or LB? > > slr > > --------------------------------------------------- > PLUG-discuss mailing list - PLUG-discuss@lists.plug.phoenix.az.us > To subscribe, unsubscribe, or to change you mail settings: > http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss > --------------------------------------------------- PLUG-discuss mailing list - PLUG-discuss@lists.plug.phoenix.az.us To subscribe, unsubscribe, or to change you mail settings: http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss