How to know the maximum load of my application or server. I need to know this to decide whether my application need to be cluster or not.
If we have to do a test to know, so we need a lot of concurrent client to hit the server. But we can't simulate this like in the real world, how to simulate concurrent millions of people..?
Maybe we can test for a single server in a production. But when it fail, it will be a bad user experience. Of course we can improve it after that.
Knowing the maximum load earlier might be better.