CLOUD WORKFLOW SCHEDULING BASED ON STANDARD DEVIATION OF PREDICTIVE RESOURCE AVAILABILITY

Abstract

Cloud computing has emerged as a new paradigm that provides services to the user on pay-as-you-use basis. Cloud computing’s Infrastructure as a service provides various resources to user such as computing, storage and bandwidth etc., where Computing services can be provided in the form of Virtual Machines. Parallel applications submitted for execution on these Virtual Machines can be completed without any delays only when the Virtual Machines are available for execution. In general, resources in cloud are shared by various users and their availability is dynamic and unpredictable due to various reasons such as capacity of underlying hardware and number of users sharing the resources and various other reasons. Hence, non-availability of these allocated resources may cause delays in completion of execution of parallel applications. In this scenario, existing algorithms without considering the delays that may occur due to non-availability of Virtual Machines may not perform better and are less reliable in terms of completion of jobs. In this paper, we propose a new heuristic called Workflow Scheduling based on Standard Deviation of Predictive Resource Availability in cloud computing considers the dynamic nature of cloud resources and its dynamic availability in scheduling decisions and produces reliable schedules.