Energy Efficient Resource Utilization and Load Balancing in Virtual Machines Using Prediction Algorithms

Abstract

In recent years due to increase in the number of users and clients adopting Cloud Computing for both personal and business needs have drastically increased. As a result of this increase in utilizing the Cloud services, Cloud datacenters are witnessed to be massive energy consumers and environmental polluters. The Primary goal of ensuring loads on virtual machines is to reduce energy utilization and providing maximum resource allocation. Balancing the loads on the servers has become a challenging task for the cloud service provider. Load balancing on servers helps to improve the performance of the virtual machines and to minimize energy and processing time in cloud systems. Here we discuss about comparison of various existing static load balancers as well as the conventional dynamic load balancer.  The problem of reducing the energy in Cloud datacenters are being addressed from various research perspectives, predicting the future trend and behaviours of workloads at the datacenters and thereby reducing the active server resources. However, this includes various practical and analytical challenges imposed by the increased dynamism of Cloud systems. The behavioral characteristics of Cloud workloads and consumers are still not defined clearly. We present a novel resource optimization framework to avail the most optimum level of resources for executing jobs with reduced server energy expenditures and job terminations. This optimization framework encompasses a resource estimation module to predict the anticipated resource consumption level for the arrived jobs and a classification module to classify tasks based on their resource intensiveness.

Keywords: Load balancing, Cloud datacenters, Energy prediction, Resource allocation, Energy efficient resource utilization algorithm.