Energy-Aware Consolidation Scheme for Data Center Cloud Applications

Abstract

The consolidation of resources is one of the most efficient strategies to reduce the power consumption in data centers. Various algorithms have been proposed in order to reduce the total number of required servers and network devices. The practice developed in response to the problem of server sprawl, a situation in which multiple, under-utilized servers (and/or network devices) take up more space and consume more resources than the ones justified by their workload; with the effect to power off unused equipment. Generally, consolidation mechanisms consider different parameters related to the services neglecting the specific function of the Virtual Machines (VMs) in the application framework (e.g., core component, backup replica, member of a set of workers for load balancing). In this work, we develop a new consolidation algorithm that takes into account the particular function of each VM with the aim to apply power saving mechanisms without compromising the desired service level. The results of the simulations show that it is possible to obtain significant energy savings. In particular, we show, with different heuristics, the optimal trade-off between service level and power efficiency achieved by the proposed model.