Cutting the electric bill for internet-scale systems

TLDR

Energy costs are a growing share of data center expenses, and electricity prices vary temporally and geographically due to demand, transmission, and generation differences. The study aims to characterize how fluctuating electricity prices affect computation costs and to show that distributed systems can exploit this variation for substantial savings. Using historical electricity prices for 29 U.S. locations and Akamai CDN traffic data, the authors simulate realistic workloads to quantify potential economic gains.

Abstract

Energy expenses are becoming an increasingly important fraction of data center operating costs. At the same time, the energy expense per unit of computation can vary significantly between two different locations. In this paper, we characterize the variation due to fluctuating electricity prices and argue that existing distributed systems should be able to exploit this variation for significant economic gains. Electricity prices exhibit both temporal and geographic variation, due to regional demand differences, transmission inefficiencies, and generation diversity. Starting with historical electricity prices, for twenty nine locations in the US, and network traffic data collected on Akamai's CDN, we use simulation to quantify the possible economic gains for a realistic workload. Our results imply that existing systems may be able to save millions of dollars a year in electricity costs, by being cognizant of locational computation cost differences.