1

There is a place in jMeter where you can specify the number of users you want to start concurrently. I assume you can't put a number where it exceeds the limitation on your hardware & OS.

http://jmeter.apache.org/usermanual/build-web-test-plan.html

If you set the value to 0, then JMeter will immediately start all of your users.

But given a specific hardware and OS, let say windows 7, i5 CPU, 8GB RAM. How do I know or calculate the number of users I can put to start all the users concurrently? Although I can trial & error to find out the numbers, I hoping if there is a way to pre-calculate it.

Sven
  • 98,649
  • 14
  • 180
  • 226
forestclown
  • 945
  • 4
  • 15
  • 25
  • To clarify my question: What I am trying to figure out here is let say I put number of threads = 500, and ramp-up time = 0. How do I know jMeter actually starts all 500 threads in zero seconds. Because I am sure each thread will take certain amount of hardware resources (e.g. memory, CPU). Because I assume if hardware is busy, some of the threads might not start immediately? – forestclown Apr 23 '12 at 02:02

2 Answers2

3

The ramp-up time is not really important except for special cases, so they recommend in the same paragraph to keep it at the default value of 1 sec, which means all threads you enter in the number of threads field will be started within 1 second. Also, for the local OS it doesn't matter at all if you start all threads at once or with a ramp-up time, the limiting factor will be the number of threads alone.

How large you can select the number of threads value on the client machine is dependend on many factors and you have to test it yourself, but since the purpose of JMeter is to see how your server handles X concurrent connections, I assume in many cases the server or the network connection to the server will be the limiting factor, not the local machine itself.

Sven
  • 98,649
  • 14
  • 180
  • 226
  • 1
    Thanks for your feedback. What I am trying to figure out here is let say I put number of threads = 500, and ramp-up time = 0. How do I know jMeter actually starts all 500 threads in zero seconds. Because I am sure each thread will take certain amount of hardware resources (e.g. memory, CPU). Because I assume if hardware is busy, some of the threads might not start immediately? – forestclown Apr 23 '12 at 01:59
  • I am sure here is no general metric to calculate this, but I doubt this will be relevant at all, except you have a very large number of threads to start and you really need an instant ramp-up, but why would you? – Sven Apr 23 '12 at 12:03
  • I am hoping to test how the server can handle sudden surge of traffic. But the test might not be accurate if I am not sure if jMeter can or cannot start X number of concurrent users within 0 seconds? – forestclown Apr 24 '12 at 01:39
0

To check all threads were started simultaneously use listeners. For instance View result in table, see 'Start time' column

Jay
  • 111
  • 1