Comparative Study of the Virtual Machine Scheduling Approaches in Cloud Environment

In the current computing and technology scenario, computing, storage, and network resources are provided as a service to the end-users. The users put the demand from different time zone using scalable simulation. The requests are handled using service-oriented computing. The requests map on virtual machines. Our objective is the detailed comparative study of the static, dynamic, metaheuristic, and bio-inspired neural methodologies for the assignment of user requests on virtual machines. The optimal scheduling of requests on virtual machines depends on the resource assignment policy defined at task placement on a virtual machine across the data center network. The comparative study covers the dependent and independent virtual machine level scheduling using a scalable simulation with user-defined policies. The comparative results show that when shifting from a static approach (FCFS) to metaheuristic scheduling methodology (BB-BC-COST), the performance metrics improve by 26.97% when the maximum number of cloudlets are mapped on virtual machines.