On Arbitrating the Power-Performance Tradeoff in SaaS Clouds
In this paper, the authors present an analytical framework for characterizing and optimizing the power-performance tradeoff in Software-as-a-Service (SaaS) cloud platforms. Their objectives are two-fold: they maximize the operating profit when serving heterogeneous SaaS applications with unpredictable user requests and they minimize the power consumption when processing user requests. To achieve these objectives, they take advantage of Lyapunov optimization techniques to design and analyze an optimal control framework to make online decisions on request admission control, routing and Virtual Machines (VMs) scheduling. In particular, their control framework can be flexibly extended to incorporate various design choices and practical requirements of a datacenter in the cloud, such as enforcing a certain power budget for improving the performance (dollar) per watt.