A Concurrent Level Based Scheduling for Workflow Applications within Cloud Computing Environment

Cost minimization under deadline-constraint-based workflow scheduling described by Directed Acyclic Graph (DAG) is a NP-hard problem in Cloud Environment. In order to address such problem, this paper proposes a novel heuristics approach of Concurrent-Level-based Workflow Scheduling (CLWS). It stratifies all the tasks according to the concurrence among tasks during the actual workflow execution. CLWS distributes the total redundancy time into every level according to their concurrent degree. As well as it adopts the algorithm of Markov Decision Process (MDP) to optimize tasks, which have time dependence with each other in the same level. The Simulation results show that CLWS can give a better optimized result.