hi guys. I have been reading these forums for a long time and I know there are other products besides eaccelerator, so the main point of my question is not necessarily related to eaccelerator but applies to any kind of tool that offers the same functionality. I am hosting several friends, no real customers on my machine running the latest ispcfg 2.xx and have been successfully using eaccelerator for a long time now. Most of the clients are using wordpress. Now I started using a caching plugin for wordpress that can cache onto disk and/or eaccelerator cache. BUt I was wondering as WP instances could easily generate over 15M of compiled scripts in eAccelerator each, so is there any real point in using eaccelerator? I mean I can't allocate 15MB cache per client to eaccelerator, that would easily be 150MB! I was running eaccelerator with the default size of 32MB and thought the TTL and Prune settings of eaccelerator woudl take care of the distribution of the cache? What do other "providers" do? how much RAM do you invest in eaccelerator? does eaccelerator distribute the available cache on a first-come first-served basis? any other thoughts about this matter are highly appreciated.
I don't know much about eAccelerator's internal workings, but I'm always using something around 30MB. The max. value is limited by what you have in /proc/sys/kernel/shmmax (see http://eaccelerator.net/wiki/Settings ). As most eAccelerator installations use something around 30MB, I'm sure the WP developers took care of that.
