thanks in advance.
Well an idea we've got a dell poweredge 1950 with raid 1 sas 73gb 8gb memory dual xeon 5430 on a Red Hat Enterprise Linux x64 Mysql 5.0.45, php 5.1.6 and Apache 2.2.3 We've got aproximadly 300 simultaneous connections in a server of 14k users and 1k courses.
if you want, i can send you the config of httpd then you can just see the scalability of the configuration.
also can i use these machine specification with about 800 - 1000 simultaneous connections.
well i can sant to you the httpd.conf and mysql configuration to see what heapens, the how to configuration deppends on your hardware i can send the apache page where i found the variables to test and see if you can apply this number of connections.
the configuration and the site goes inside the file (donno if i can post links to other sites here) :D
Anyways i hope that it helps you there
i will try it
man i'am signed to the apache comunity and i've recived this message when talking about connections in apache
Anthony J. Biacco
1. I have to rail totally against this. The more you lower MaxRequestsPerChild, the more often apache is killing and recreating a child process. At numbers as low as 2000 or lower, you’re starting to defeat the whole purpose of using the worker mpm.
>=50% of apache’s time is going to be spent managing child processes on a high traffic site. MaxRequestsPerChild should either be 0 or something very high. IF your process memory usage gets higher and higher, then you have a memory leak somewhere.
2. Don’t use ThreadLimit, stick with ThreadsPerChild
3. You MaxClients doesn’t sync up to your other numbers. MaxClients is going to be ServerLimit x ThreadsPerChild. So for you, 1500. If you want to serve 1500 concurrent reqs, then set MaxClients to match this at 1500. If you want 500, then change ServerLimit,StartServers and ThreadsPerChild so the math is right. For instance, ServerLimit 10, StartServers 5, ThreadsPerChild 50 will be you a MaxClients of 500.
If you give us your server parameters (cpu, memory, modules loaded, apache rss usage, types of files served), we’d be able to better recommend numbers for what your server can support.
Manager, IT Operations
Format Dynamics, Inc.
I Hope it help's you cause i didn't try this still i'll get along with him
In case of web aplications like Moodle crucial is 'simultaneous users' parameter. I thnk that would be cheeper to buy two smaller servers than single big one. Dell PoweEgde 1950 mentioned before would be great but I would consider expanding RAM up to 16 GB on both.
When talking about DB part, key factor is disk I/O so 4 disks would be better than 2 (more heads, shorter RandomSeek time).
When building system that scale, IMHO you should consider a few more factors than only performance.
First one is reliability. How quick should you recover system after crash? (MTTR). How many data lost during crash would be still acceptable?
I think you should first estimate those factors and then plan system architecture. Hardware should be adjusted to it.
If you keep on increasing CPU power any memory size, won't you hit the data i/o speed of the permanent storage (harddisk or NFS or SAN or whatever)?
If you get that boosted, won't you hit the speed of the network interface, current level 1 GBit/s. If your server is utilized by 5% of the users, that gives 1000 MBit/s : 5000 users = 200 kBit/s per user.