Just want to double check and confirm if this is correct or not. I have 3 scenarios each using per-vu-iterations and in each scenario there are 30, 60 and 150 vus with 1 iteration.
When I run the test with all 3 scenarios, the console shows that 210 VUs are created.
See the startTime: 2m15s of the last scenario. The first scenario starts immediately and only takes 1m30s at most (maxDuration: 1m0s + gracefulStop: 30s). So k6 knows it can reuse the VUs from the first scenario into the third one, and allocates only 210 VUs for better performance and efficiency.