Bi-objective HWDO Algorithm for Optimizing Makespan and Reliability of Workflow Scheduling in Cloud Systems

Cloud computing is hyped in a computing enhancement by enabling the user to access dynamically scalable services through Internet based on pay-per-use model. The infrastructure evaluated as a platform for executing scientific workflows with resource elasticity and heterogeneity. Considering the failure of the resource along with time and cost attribute become important while allocating resources to the tasks in these workflows. The objective of the present study is to design a algorithm to optimize both makespan and reliability. In this paper, we presents a bi-objective algorithm based on wind driven optimization(WDO). A mutation operation is hybrid with the working of WDO (HWDO) algorithm that attempts to generate the schedule with maximum reliability. The goal of proposed algorithm is to approximate the solution to the scheduling problem that depicts the trade-off between minimizing makespan and maximizing the reliability of the schedule. The algorithm is evaluated on real world scientific workflows in cloud based simulation environment WorkflowSim. The results depicted an improvement in optimizing the stated parameters in contrast with other studied algorithms.