Virtual Machine Consolidation Framework for Energy and Performance Efficient Cloud Data Centers

Abstract

Establishing large-scale data center increases demand in computational power of cloud computing.. So cloud datacenters extensively applied a virtualization technology which is realizing energy efficient operations of cloud datacenter. It is a method of an abstraction of the computing resources to make increases the utilization of resources. Virtual machine (VM) consolidation is a method of combining workloads from separate machines or applications into a smaller number of systems. Consolidation takes place to run the maximum number of VMs on a single Physical Machine (PM) to increase the performance of the PM and also to reduce the total power consumption of the data center. While, live migration process PM’s ruining virtual machine is transferred to another active PM, so that source PM could be switched off to reduce the energy consumption. Both technologies, developing a data center that provides a good tradeoff between the performance and energy efficiency. There are three contributions in the framework. First, PM overload detection based on Service Level Agreement Violation (SLAV) and under load detection based on minimum number of VM for VM allocation algorithm is established. Second, in VM selection algorithm VM with maximum utilization for migration has been selected. Third, VM placement algorithm using Minimum Power High Available Capacity algorithm for finding new places of VM has been developed. We have evaluated our framework in the CloudSim simulation tool. Then investigated performance metrics of VM consolidation and migration. Experimental results show an effective VM consolidation and migration strategies can improve the energy efficiency.