improving tradeoff between energy and performance metrics in resource management

Rapid growth of the demand in cloud infrastructure has led to the establishment of large-scale virtualize data centers in all around the world. Such data centers consume huge electrical energy, which has a negative impact on the environment, carbon dioxide emissions and high operational costs in context conservation and services presentation issues in cloud. Also, energy consumption is transformed to very significant subject in resource management process. Therefore, cloud providers are try to reduce their energy consumption while ensuring requiredQuality of Service (QoS) based on the Service Level Agreement (SLA). In this paper, we propose a new procedure of Virtual Machines (VMs) management based on a multi-criteria method to reduce energy consumption while meeting the QoS requirements.Due to aggressive consolidation which may lead to performance degradation from QoS perspective, we present policies for both determination of under-loaded servers and also VMs placement in consolidation and/or de-consolidation process into VMs management procedure. The simulation results through CloudSim show that the proposed policies noteworthy results concerning energy-performance tradeoff with reductions in terms energy consumption, SLA violation, and number of VM migrations

Keywords:

Resource Management , VM Placement , underloaded hosts , SLA , Energy-efficient