Running benchmarks?

[Anon]

I have two versions of a script I have made, and it's not clear by looking at it which is the most server friendly to run (resources, speed). How would I go about testing for this? If it takes any special programs, I'm running Win98... but could install Linux if I absolutely have to.

Thanks.

[cordex]

microtime();

I'm sure you can figure out the subtraction involved. grin

[jariizumi]

If you really want to get good results, you can install Linux or get a hold of a Linux server or a Win32 version of AB (ApacheBenchmark), and do something like

ab -n 20000 -c 20 http://www.mysite.com/script1.php

then compare with

ab -n 20000 -c 20 http://www.mysite.com/script2.php

where 20000 is the number of requests and 20 is the number of simultaneous requests.

Measuring time by microtime is good, but doesn't take into account the time to server/database caching, repeated access, etc. Hope this helps 🙂