I've read and heard that nods.js should be better and lighter at the server when doing long-polling then on a PHP-server. I'm doing a study on this and would like to measure the serverload at two simple long-polling applications to see the difference.
How could I measure this in a good way? I read something about apache bench but don't know if that works when doing long-polling. Also, I'm running on localhost.
Here are my two applications if someone want to take a look: https://github.com/furst/longpolling