I’d like to test the performance of the system under the load from different users during a long period of time. For example, I’d like to run a test for 10 hours and the test should use 1k users. The catch is that I’d like the users to be different and rotate each 30 minute or so. So, 1k users spawn, do their thing for 30 minutes, then they start despawning and another 1k spawns to run for the next 30 minutes.
What I could do is use stages, something like this:
-s 10s:100, 1800s:100, 10s:0, 10s:100, 1800s:100, 10s:0... //repeating for the desired number of batches
My question is, is there any way to elegantly setup the test with the desired properties or is the repeating stages configuration the best possible way?
The scenario also looks a bit similar to this one: https://k6.io/docs/using-k6/scenarios/advanced-examples/#using-multiple-scenarios
However, in my case I don’t have multiple different scenarios, but rather N copies of the same scenario.