Energy Efficient Virtual Machine Consolidation in Cloud Data Centers

Abstract

With the tremendous increase in Internet capacity and services, the demand for cloud computing has also grown enormously. Cloud computing provides on-demand access to computing resources for users across the world. It offers services on a pay-as-you-go model through data center sites scattered across diverse geographies. However, cloud data centers consume a massive amount of electricity and leave a high carbon footprint in the ecosystem. It has been observed that the electricity consumption of data centers is around 2% of the total electricity consumed all around the world, and it is rising by 12% every year. Also, these data centers are responsible for producing 2% of the global CO2 emission, which is the same as the global aviation industry. As a result, having energy and carbon-efficient strategies for remote cloud data centers is unavoidable. Reducing energy consumption while maintaining service level agreements (SLAs) is one of the most critical issues in this optimization effort. An essential source of energy waste in cloud data centers lies in the inefficient usage of computing resources. newlineAccording to existing research on power consumption in cloud data center indicates the idle power in servers is always above 50% of peak power. Therefore, keeping server underutilized is very inefficient from the energy consumption point of view. Hence, energy consumption in the cloud data center has been indented as one of the critical research challenges in recent times and many researchers have been working in different directions to address the issue. newlineThe virtual machine (VM) consolidation process has been selected to improve energy consumption by minimizing the number of active physical hosts. The process of work- load consolidation includes selecting a Virtual Machines (VMs) from overutilized hosts and trying to put them on other hosts such that this source host becomes normal and the target hosts do not get overutilized.