0

I have concerns about the duration of executing JMeter Test Scenario. In fact, if we execute a manual test with 1 user in 1 hour, we will find the same duration or little more with 3 user. But, with JMEter test the duration with 1 user will be multiplied by the number of users.

During my scripting with this tool, I've noticed that JMeter always wait for the response of the request to pass to another request. It's like we have 1 user doing the work of multiple users.

Does any body have explanations about this issue? Can we configure JMeter to perform like we have x users working in // ?

Improve this question

2 Answers 2

Reset to default
1

As far as I know jmeter doesn't work in a sequential order. What jmeter does is, it takes all the requests and send them to the server/host all at the same time. So if you have configured it for 3 users making 1 http(s) request, that is 3*1 http(s) requests will be sent to the server/host all simultaneously and then it will wait for the response!

Improve this answer
6
  • Thanks for your response, this is exactly what i have learned about JMeter during Load Testing. But, i still have doubts about the duration with 1 user and more.
    – Imen CHOK
    Commented Oct 6, 2014 at 7:11
  • What jmeter or any other tool for that matter does for you is create a virtual environment which may or may not be exactly the same as the actual real life environment. So you can always get an idea of what things will be like in production without any certainty. So the time required will always depend on the load you are generating and the way in which the tool will generate it. You will have to learn more about how jmeter works and also the environment and project that you are testing using jmeter.
    – IAmMilinPatel
    Commented Oct 6, 2014 at 7:22
  • This is exactly what i'm trying to do, learning more about the application under test and how JMeter really works so that i can defend my load tests. PS: your interpretation are so logic. In fact, JMeter will simulate things like in production without any certainty.
    – Imen CHOK
    Commented Oct 6, 2014 at 7:29
  • Thats right, so if anyone wants to know for certain how the application and server will react in real time load, well the only way is to monitor it in real time. Otherwise we can use tools and try to get as close as we can to the real environment but never actually clone it to perfection! Hope whoever you are trying to explain this does infact really get it! :)
    – IAmMilinPatel
    Commented Oct 6, 2014 at 7:33
  • Thanks again for your help, i will let you know about arguments that I will use to convince them that i'm using the right load test tool :)
    – Imen CHOK
    Commented Oct 6, 2014 at 7:46
0

The options to precisely set desired throughput are in:

  1. Ultimate Thread Group (available via JMeter Plugins) provides easy visual way of defining virtual users ramp-up, ramp-down and time to hold the load. 3 users per hour will look like

Ultimate Thread Group

  1. Constant Throughput Timer allows to set target load in "requests per minute" so if you want 3 requests per second put 180 in "Target Throughput" input. In this case test duration can be limited via Thread Group "Scheduler" functionality or using a parent Runtime Controller

Constant Throughput Timer

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.