Battery Matters: Rightsizing Energy Storage for

Abstract—Modern cloud datacenters are equipped with mul-tiple energy sources including smart grids, battery and on-site renewable energy. The battery places an important role in storing renewable energy and power purchased from the grid markets, for green and cost-effective energy consumption. There is a lack of study in the existing literature on quantifying the role and cost tradeoff of rightsizing the battery for energy store in a cloud datacenter. In this paper, our goal is to obtain a thorough understanding on the tradeoffs between battery’s contribution in energy storage and its provisioning cost, in order to gain useful insights for battery rightsizing in cloud datacenters with arbitrary energy demand and time-varying renewable energy generation. We further consider different pricing schemes of the smart grids, including constant pricing, dynamic real-time pricing and two-timescale pricing. Under each pricing scheme, we formulate the problem as a stochastic optimization model that minimizes the long-run operational cost of a datacenter, and derive adaptive, optimal control policies that guide power purchase from the smart grid and battery charge/discharge efficiently. We also demonstrate how the cost and efficiency of the battery, pricing schemes and system uncertainties impact the management and optimal capacity of the battery, with both theoretical analysis and trace-driven experiments. I