If billing is based on 'hours of consumption', how is that consumption metered? If a customer had, for example, 64 physical cores powered on and assigned to the workload (128 users), providing 128 vCores, with a light load of 1vCore/user with an average of 260 hours active work across the month*, would they be charged 260 hours x 128 cores x $0.01 per month ($333/month) or would it be based on the full 720 hours ($921/month) since on-prem workloads don't tend to be powered down as aggressively as cloud hosted workloads.
*Likelihood is that it would be more than this with multi-session workloads even if machines are AutoScaled efficiently due to session lingering etc.
Also, just to be clear, is the $0.01/vCore/Hour fee in addition to the $10/physical core/month Azure Stack Host Service fee (so in this case, an additional $640/month)
Thanks
Danny