Optimal calculation overhead for energy efficient cloud workload prediction

Abstract

Amazon recently estimated that the cost of energy for its datacenters reached 42% of the total cost of operation. Our previous research proposed an algorithm to predict how much cloud workload is expected during a future time interval. Accurate knowledge of the future workload allows the datacenter operator to place unneeded physical servers in a low-power state to save energy. If more system capacity is required, servers in a low-power state are transitioned back to an active state. In this paper, we extend our prior research by presenting a new approach to determining the frequency of calculating the prediction of the expected capacity. We present a dynamic prediction quantization method to determine the optimal number of prediction calculation intervals. These new algorithms allow us to predict future load within required Service Level Agreements while minimizing the number of times the prediction calculations must be performed. We finally test this model by simulating the stochastic time horizon and dynamic quantization algorithms and compare the results with three competing methods. We show that our model provides up to a 20% reduction in the number of calculations required while maintaining the given Service Level Agreement.