Power-Consumption Aware Algorithm for Resource Allocation in Cloud Computing
Cloud Computing is an emerging technology. Cloud Computing provides many facilities like on demand self-services, unlimited resources, rapid elasticity and measured services to end users. All users access these resources directly through the Internet. Users can use these resources and services as they want, on pay per use concept. In cloud computing architecture, load balancing is an important issue. There are many algorithms for resource allocation in cloud computing, which work in different ways. We propose a technique to allocate resources to a cloudlet, by taking into considerations of not only the resource capability but also the estimate of the energy to be consumed, whilst serving the cloudlet. In the proposed approach, the there is a provision of matching energy consumption as a criteria while making a selection of VM is optional by the cloud user to participate in energy-efficient resource allocation. It is based not only on the execution time or completion time as a selection basis, but also estimated power consumed by the data center. The main motive of our work is to balance the total load of cloud system, so as to achieve higher energy efficiency. We simulated results using the CloudSim toolkit. Results show the comparison of energy consumption, in performing the same set of tasks, with reduced power consumption power consumption .
Keywords - Cloud computing, load balancing algorithms, Service Level Agreement, Cloudlet, DVFS, QoS