How to know how many requests my jmeter can send maximumly per second

Question

I could use ConstantThroughputTimer to control the number of requests per second in jmeter. If the request number is very big, though I set the number, I couldn't get what I want.

For example, I'd like to send 10000 request per second and I've set the request number, but the computer which running my jmeter can't generate 10,000 requests per second due to its hardwares. Maybe it only can generate 1,000 requests per second maximumly. If the throughput of the target webapp is 1,000/sec, I can't determine the cause of the test result. Both the webapp and my jmeter may be the bottleneck.

So I'd like to know the maximum requests my jmeter can send.

Answer 1

It depends not only on your JMeter machine hardware specifications, but also on what your test is doing, request and response size, number of pre/post processors, assertions, etc. so the number may vary depending on the nature of the test as JMeter will wait for response from the previous request prior to starting the next request.

1. Make sure you have the optimal JMeter configuration, i.e:

   • increase JVM Heap allocated to JMeter
   • remove all the listeners from the test plan
   • run your test in non-GUI mode
   • follow other recommendations from the 9 Easy Solutions for a JMeter Load Test “Out of Memory” Failure guide
2. Constant Throughput Timer is only capable of pausing threads, it will not kick off extra virtual users to reach the target throughput so make sure you provide enough Threads (virtual users) on Thread Group level
3. If you won't be able to reach the desired throughput using single JMeter instance you will need to consider Distributed Testing