normally we use litespeed in complicated crazy high load environments. but I happened to be doing a plain old simple cpanel install and decided to use litespeed there also because we use suphp.
im used to installing litespeed and configuring it myself so I was prepared to go through the setup, change all the options i like, install dependencies, prepare and compile php myself/etc.
but instead I followed instructions for the cpanel plugin. 10 seconds later litespeed is installed. one click + 5minutes later php is compiled and installed. next click its swapped from apache to litespeed.... and im done. I was impressed how easy it was.
because its so easy to swap back I did some quick apache bench tests.
I did not touch the server config at all, just wanted a quick baseline comparison with nothing special or magical done.
tests were done from another server 0.6ms away on a 10mbit internet connection. suphp is enabled. I did several tests but included only one result here, the results were similar no matter what test i tried.
test.php:
apache:
Requests per second: 406.75 [#/sec] (mean)
observed load average: 1.61
litespeed:
Requests per second: 1977.87 [#/sec] (mean)
observed load average: 0.28
1/5th the load, 5x more requests per second.
is a nice result to see.
browsing to the forum and writing this post took more time than all of the above combined
im used to installing litespeed and configuring it myself so I was prepared to go through the setup, change all the options i like, install dependencies, prepare and compile php myself/etc.
but instead I followed instructions for the cpanel plugin. 10 seconds later litespeed is installed. one click + 5minutes later php is compiled and installed. next click its swapped from apache to litespeed.... and im done. I was impressed how easy it was.
because its so easy to swap back I did some quick apache bench tests.
I did not touch the server config at all, just wanted a quick baseline comparison with nothing special or magical done.
tests were done from another server 0.6ms away on a 10mbit internet connection. suphp is enabled. I did several tests but included only one result here, the results were similar no matter what test i tried.
test.php:
PHP:
<? print "hello"; ?>
Code:
ab -c 20 -t 120 -n 200000 http://.../test.phpobserved load average: 1.61
litespeed:
Code:
ab -c 20 -t 120 -n 200000 http://.../test.phpobserved load average: 0.28
1/5th the load, 5x more requests per second.
is a nice result to see.
browsing to the forum and writing this post took more time than all of the above combined