An Efficient Virtual Machine Scheduling Algorithm with Heterogeneous Tasks for Cloud Computing Environments

Abstract

Cloud computing is a technology that provides access to the services anywhere, anytime and paying for the services that are being used. This allows cloud users to use the data and application if needed anywhere. The main task to be handled is the services arriving at the server for service. In cloud computing environment to get services, Virtualization is the significant feature to be concerned. Which allows multiple operating systems can run simultaneously on a single physical machine. In cloud environment number of user can request for the services simultaneously, so there is a need for scheduling mechanism that efficiently schedules the resources to the specific requirements of the users. In the other way in cloud computing environments there are two role players, the cloud service providers and the cloud service users, with having their own goals of requirements. In the case of service providers need scheduling algorithm, which maximizes their profit and cloud service providers need to execute with less time in getting the quality of services. In this thesis, we had proposed An Efficient Virtual Machine Scheduling Algorithm with Heterogeneous tasks on cloud computing environments. To achieve this goal, we had reviewed the existing scheduling in cloud computing and we have identified some drawback of researches done before. After this, we mainly focused on design and implement scheduling algorithm and analyzed all the existing scheduling algorithms. Then to get experimental results and to implement our proposed scheduling algorithms, we have used CloudSim simulation tools. After all we have evaluated the performance of the proposed scheduling algorithm with two existing scheduling algorithms. Mainly we have evaluated our algorithm with round robin and first come first serve scheduling algorithms. To evaluate this proposed scheduling algorithm we have used three optimization metrics, maximized average resource utilizations, reduce completion time and average response times. Based on these optimization metrics the proposed scheduling algorithm performs better than both Round Robin and First Come First Serve scheduling algorithms.