The k6 perf test results fluctuate a lot across different runs for release 0.107. Close examination shows it's largely due to k8s schedule dynamics that in every run, the speed to ramp up pods, and the actual pod CPU usage level vary a lot. Thus, we need to adjust resources configuration and HPA to get consistent resources thus consistent test results.
Steps to reproduce
Run the k6 test suite multiple times and compare the results.
Description
The k6 perf test results fluctuate a lot across different runs for release 0.107. Close examination shows it's largely due to k8s schedule dynamics that in every run, the speed to ramp up pods, and the actual pod CPU usage level vary a lot. Thus, we need to adjust resources configuration and HPA to get consistent resources thus consistent test results.
Steps to reproduce
Run the k6 test suite multiple times and compare the results.
Additional context
No response
Hedera network
other
Version
v0.107.1
Operating system
None