Energy Efficient Distribution of Heterogeneous Workloads in Cloud Data Center

Cloud computing has become a popular technology for on-demand services for end consumers. Geo-distributed data centers conduct most cloud computing jobs and may use a lot of energy. In this paper geo-distributed cloud data center energy-efficient workflow scheduling approaches are discussed. Scheduling jobs to virtual machines to prevent overloading servers is a key scheduling goal. To address the scheduling issue, load balancing is necessary. Effective load balancing reduces response time and maximizes resource use. In this paper, a technique for energy-efficient workload assignment to cloud data centers is proposed. The heterogeneous workloads from Google dataset are classified using machine learning techniques for energy-efficient assignment to appropriate data centers. Various clustering algorithms produced equal clusters. In our proposed approach, 2 clusters segregated data properly. The performances of various load balancers in varying scenarios are also compared in this research.