A distributed speed scaling and load balancing algorithm for energy efficient data centers

Modern electronic communications rely on large-scale data centers to process and store the vast amounts of data produced by smart phones, tablets, and other mobile devices. Remarkably, most data centers are still not energy efficient, because they operate full speed around the clock regardless of demand. In this paper, we present a novel distributed algorithm that adjusts server speed and workload routing according to the demand of multiple (virtualized) applications. Unlike previous studies that use the mean response time of a simple M / M / 1 queue, the proposed algorithm incorporates the response time distribution obtained from a G / G / 1 / P S queue. We explain the optimality of the proposed algorithm and discuss the impact of incorporating response time constraints. The numerical results show the effectiveness of the algorithm when implemented in a distributed manner.