Combined Power and Performance Management of Virtualized Computing Environments Serving Session-Based Workloads

This paper develops an online resource provisioning framework for combined power and performance management in a virtualized computing environment serving session-based workloads. The authors pose this management problem as one of sequential optimization under uncertainty and solve it using Limited Look-a-head Control (LLC), a form of model-predictive control. The approach accounts for the switching costs incurred when provisioning virtual machines and explicitly encode the risk of provisioning resources in an uncertain and dynamic operating environment. They experimentally validate the control framework on a server cluster supporting three online services. When managed using LLC, their cluster setup saves, on average, 41% in power-consumption costs over a twenty-four hour period when compared to a system operating without dynamic control.