Enhancement in Cloud Performance using the Clustering Method of Scientific Workflow Tasks

In the fast-paced competitive environment, cloud computing is developing satisfactory interest in the community of scientific applications. For executing complex data sets of scientific applications, scientific workflows are widely used by various researchers. If we talk about the scheduling of tasks of such workflows on the cloud then cloud resources are required. Here, the requirement is how we schedule millions of billions of tasks of a workflow so that cloud performance can be enhanced. It is highly recommended to work on ranking and clustering immediately after downloading the data sets of input tasks. The work aims at maximizing resource utilization while accepting deadline constraints of tasks. Only to have the intention of minimizing the execution time, can create the problem of underutilization of resources. Therefore, it is mandatory to think about various factors which can support fewer time requirements for execution but side by side also take concern the more resources utilization. In this paper, we have proposed a new method called Ranking and Clustering (RCM) to provide a grouping of ranked tasks before scheduling.