Generate an artificial workload to test stability #44
Description
At the moment we verify that we do not crash by running CamFlow for a while in a VM. It helps catch some issues, but we need a more systematic approach and a more "representative" workload. Running benchmark? Look at some artificial workload framework? Stress test?