By default, Apache comes preconfigured to serve a maximum of 256 clients simultaneously. This particular configuration setting can be found in the file /etc/httpd/conf/httpd.conf
If your server has 2 GB of RAM, and you’re sharing your server with MySQL(true in my case), you’ll want to reserve about half of it for Apache (1 GB)
MaxClients: here is the process of determining MaxClients. type
ps -U apache -u apache u
See the number of apache process running in you command prompt.
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
apache 7694 0.0 0.3 42704 6680 ? S 18:30 0:00 /usr/sbin/httpd
The above indicates that a single httpd process is using 6.6 MB of RSS (Resident Set Size) memory (or non-swapped physical memory) and that it is using 42 MB of VSZ (Virtual Size) memory. This depends on the number of modules you have loaded and running in Apache.
As shared libraries are included in this number, it’s not 100 percent accurate. We can assume that half the RSS number is “real” memory. Let’s assume that each httpd process is using (6.6/2=3.3) 4 MB of memory. So if you have 1 GB ram then divide it with 4 MB of memory, which leaves room for around 256 concurrent httpd processes. Set MaxClients 256
Or
Somebody prefers to set MaxClients using following rule
MaxClients = 150 x RAM (GB)
So for example if you have 2 GB RAM (dedicated for apache) set this value to 300. In my case IT WILL BE 150
Or
Some individuals maintain that each httpd thread uses about 5 MB of “real” memory. So they determine by the following way..
Or MaxClients = RAM(MB)/5
So for example if you have 2 GB RAM (dedicated for apache) set this value to 409. In my case IT WILL BE 204(1 GB for apache)
Note: There is no reason for you to set it any higher unless you have a specific problem with this value. A high value can lead to a complete server hang in case of a DOS attack. A value too low can create timeout problems for your clients if the limit is reached
StartServers - Sets the number of child server processes created on startup. This setting depends greatly on the type of webserver you run. If you run low traffic websites on that server set it low to something like 5. If you have resource intensive websites on that server you should set it close to MaxClients.
MaxRequestsPerChild - Controls the number of request the a child serves before the child is killed. This should not be set too low as it will put an unnecessary load on the apache server to recreate the child. I suggest setting it to 1000. But we are going to use 2000 for handling heavy traffic load properly.
MinSpareServers and MaxSpareServers - MaxSpareServers and MinSpareServers control how many spare (unused) child-processes Apache will keep alive while waiting for more requests to put them to use. Each child-process consumes resources, so having MaxSpareServers set too high can cause resource problems. On the other hand, if the number of unused servers drops below MinSpareServers, Apache will fork. Leave those values to: MinSpareServers 5 MaxSpareServers 10
ServerLimit: Its better to keep Server limit same as the value of MaxClients.
MaxRequestsPerChild: I’ve Kept default apache value for this one.
So few changes need to be made in httpd.conf file which is located in /etc/httpd/conf/ directory
<IfModule prefork.c>
StartServers 140
MinSpareServers 5
MaxSpareServers 10
ServerLimit 150
MaxClients 150
MaxRequestsPerChild 4000
</IfModule>
[Note]: Response time depends on MaxClients. If you increase the MaxClients number, server will response more quickly for each request but a high value can lead to a complete server hang.
Ab is a tool for benchmarking the performance of your Apache HyperText Transfer Protocol (HTTP) server. It does this by giving you an indication of how many requests per second your Apache installation can serve.
uptime command in your root login should not yield a load average above 1, and the server should respond to commands quickly
ab -n 10000 -c 200 -k http://your_url
-c = concurrent connections
-t = time limit
-n = # of requests
Keep tuning until you hit your maximum desired load average. For servers used interactively often, having a load above 3 is way too much to use the server comfortably. For servers used mostly as real servers, a maximum load average of 10 should be acceptable. More than that, and you’ll find yourself needing to reboot the server when experiencing heavy traffic conditions, because no terminal or remote console will respond quickly to commands, and managing the server will be impossible.
How to configure few things in php.ini file for supporting huge traffic
* Enable the compression of HTML by putting in your php.ini:
output_handler = ob_gzhandler
** Switch from file based sessions to shared memory sessions. Compile PHP with the –with-mm option and
set session.save_handler=mm
Configure mysql. Change my.cnf file for better performance.
The database parameters are tuned for systems with 1 GB RAM (for ISO CD images). If you have higher RAM, please change the following in the “my.cnf” MySQL configuration file under /etc/mysql or /etc directory.
For a machine running with 512 MB of RAM, you can set these to:
key_buffer=128M table_cache=1024 sort_buffer=64M read_buffer=2M record_buffer=4M
For a machine running with 1 GB of RAM, you can set these to:
key_buffer=256M table_cache=2048 sort_buffer=128M read_buffer=2M record_buffer=8M
For a machine running with 2 GB of RAM, you can set these to:
key_buffer=512M table_cache=3072 sort_buffer=256M read_buffer=2M record_buffer=8M
For a machine running with 4 GB of RAM, you can set these to:
key_buffer=1G table_cache=4096 sort_buffer=512M read_buffer=2M record_buffer=8M
