Energy Saving Task Consolidation Technique in Cloud Centers with Resource Utilization Threshold

Abstract

The data centers are the world’s biggest consumers of electricity. The consumption of energy in the cloud is proportional to the CPU utilization of the virtual machines (VMs). As the size of the cloud infrastructure increases the complexity of the resource allocation problem increases and becomes very difficult to solve it efficiently. This is an NP-Hard problem. There are several heuristics that may be used to solve the problem. Through task consolidation, we can get many benefits such as maximizing cloud computing resource, utilization of resources in a better way, efficient use of power, customization of IT services, Quality of Service, and other reliable services, etc. We find from the literature review that there is a high level of coupling between energy consumption and resource utilization. This paper presents the resource allocation problem in cloud computing with the objective to minimize energy consumed in computation. The simulation results show that a 70% principle of CPU utilization is the most energy efficient threshold for task consolidation in a virtual cluster. It has been verified with MaxUtil and ECTC (Energy Conscious Task Consolidation) algorithms.