Enhancing Task Scheduling in Cloud Computing with an Improved PSORL Algorithm

Cloud computing is a popular infrastructure for conducting operations with virtual computers. Task scheduling is one of the most basic problem associated with cloud computing. Makespan is the entire amount of the time needed to complete all of the scheduled jobs so optimizing makespan is the primary goal of task scheduling. Existing system includes nature inspired methods, Traditional methods like Round Robin, Shortest Job First, First Come First Serve were used to reduce makespan but found to lack efficiency. Traditional methods was implemented and compared with PSO and found that PSO’s makespan is comparatively better, but PSO could not adapt to the dynamic nature of cloud environment. Therefore to overcome the problems, the paper proposes a hybrid method using PSO and RL. Implementation of PSORL based task scheduling method to optimize the makespan. The hybrid method determines optimal mapping of tasks to virtual machines across multiple datacenters. This solution ensures in reducing the makespan.