Hi there,
I'm glad you are one step closer (or hopefully done the final step ) towards resolving the issue. It makes sense now that the "power efficient" (ie. underclocked system) could cause the VMs to be starved for resources.
Fingers crossed!