| Summary: | In the last few years, workflows are becoming richer and more complex. Workflow scheduling management system to be robust, flexible with multicriteria scheduling algorithms. It needs to satisfy the Quality of Service (QoS) parameters. However, QoS parameters and workflow system objectives are often contradictory. In our analysis, we derived an efficient strategy to minimize the overall processing time for scheduling workflows modelled by using Directed Acyclic Graph (DAG). We studied the problem of workflow scheduling that lead to optimizing makespan and reliability. The proposed algorithm handles unsuccessful job execution or resource failure by dynamically scheduling workflows to available resources. Based on the experiments results, our proposed Failure-Aware Workflow Scheduling (FAWS) Algorithm can significantly optimize the makespan and minimize the reliability by rescheduling the failed task to the unused resources. The effectiveness of the FAWS algorithm was validated based on a simulation-driven analysis based on the workflow application. © 2018 Institute of Advanced Engineering and Science. All rights reserved.
|
|---|