Deadline and Budget Constraint Based Task Scheduling Approach Using Modified PSO

Cloud computing enables users to access the online resource over the internet. Therefore, the efficient task scheduling approach should be designed to fulfil the user’s demands and achieve quality of service(QoS). In this paper, we present an improved task scheduling using particle swarm optimization (PSO) that aims to enhance the QoS. The tasks are modeled as discrete entities with specific QoS demands, and resources are specified by their distinct capacities to meet those demands. The proposed algorithm combines exploration and exploitation strategies to efficiently search the solution space based on constraints given for deadlines and budgets. Our contributions in this research will include the modification of an objective function that contains cost and execution time as the decision criteria. Our proposed algorithm DBPSO achieves 202 ms of makespan while DPSO obtains 233 ms of makespan for 250 tasks and 5 virtual machines.