Resource and Task Clustering based Scheduling Algorithm for Workflow Applications in Cloud Computing Environment

Abstract

Cloud computing consists of distributed resources and used to provide services to the applications which require huge computation power such as scientific, mathematical, weather forecasting, and biomedical applications. These applications are considered as workflow applications containing many numbers of dependent tasks. The scheduling of these dependent tasks on distributed resources is a critical problem in cloud computing. In this paper, we present a scheduling algorithm that considers clustering of resources and tasks for workflow applications in the cloud computing environment. The given algorithm is an enhancement toHySARC algorithm. Like HySARC, the proposed algorithm first forms clusters of resources and tasks and then applies list scheduling techniques on each of the clusters to schedule tasks. We have estimated and compared the performance of the proposed algorithm with HySARC algorithm on the parameters like clustering time, and makespan, and found that the proposed algorithm performed better than the compared algorithm.