It might be a good idea, to be able to simulate more than one execution controller, since on a typical HEP system also more than one WMS or user will start jobs and workflows on a typical computing infrastructure. These actors also might be spatially distributed, so they might use different entry points or to the system's infrastructure, i.e. different schedulers.