An Energy Efficient Resource Scheduling Approach for Cloud Data Centers

Abstract

Cloud Computing provides the mechanism of delivering application as services as well as the resources in data centers that provide those services. Cloud Computing has revolutionized the Information and Communication Technology (ICT) industry. These services are flexible in terms of their usage i.e., pay-as-you-use. The mapping of best resources remains a complex job in cloud environment due to heterogeneity of various resources. Scheduling the best resource–workload efficiently agreeing to user service requests is an energy optimization issue. The foremost objective of the Resource scheduler is to schedule the resources effectively and with maximum resource utilization. Resources dispersion, heterogeneity, uncertainty is a big issue for resource scheduling techniques in Cloud environment.

Current cloud computing framework hosts millions of physical servers that generate lot of heat requiring cooling units in turn to eliminate the effect of heat. Thus, overall energy consumption of the data center increases tremendously servers as well as cooling units. However existing resource scheduling techniques works mostly for virtual cloud environment. In this thesis, an energy efficient approach has been presented which schedules heterogeneous resources on the physical machines and executes cloud workloads on corresponding resources. The proposed approach improves the energy and resource utilization along with reducing the Service Level Agreement (SLA) violation. In CloudSim toolkit, we are executing proposed technique, and experimental results show that the proposed technique has better energy utilization as compared to existing resource scheduling approaches.