Power and Performance Management of Virtualized Computing Environments Via Lookahead Control

TLDR

Large‑scale data centers hosting online services are increasingly pressured to reduce power consumption, and virtualization enables consolidation of multiple services onto fewer servers while maintaining performance isolation and quality‑of‑service. This work implements and validates a dynamic resource‑provisioning framework that treats the allocation of virtual machines as a sequential optimization problem under uncertainty, solved via a lookahead control scheme. The framework incorporates switching costs and explicitly models the associated risk within the optimization objective. Experiments with the Trade6 enterprise application demonstrate that the controller reduces power consumption by an average of 26 % compared with a system lacking dynamic control, while still meeting QoS targets.

Abstract

There is growing incentive to reduce the power consumed by large-scale data centers that host online services such as banking, retail commerce, and gaming. Virtualization is a promising approach to consolidating multiple online services onto a smaller number of computing resources. A virtualized server environment allows computing resources to be shared among multiple performance-isolated platforms called virtual machines. By dynamically provisioning virtual machines, consolidating the workload, and turning servers on and off as needed, data center operators can maintain the desired quality-of-service (QoS) while achieving higher server utilization and energy efficiency. We implement and validate a dynamic resource provisioning framework for virtualized server environments wherein the provisioning problem is posed as one of sequential optimization under uncertainty and solved using a lookahead control scheme. The proposed approach accounts for the switching costs incurred while provisioning virtual machines and explicitly encodes the corresponding risk in the optimization problem. Experiments using the Trade6 enterprise application show that a server cluster managed by the controller conserves, on average, 26% of the power required by a system without dynamic control while still maintaining QoS goals.