Scheduling Independent Tasks on Heterogeneous Distributed Computing Systems Using Multiobjective Optimization Approach on Multicore Processors

The problem of mapping tasks and communications onto multiple machines and networks in a heterogeneous computing environment has been shown to be NP hard. Therefore, the development of heuristic techniques to find near optimal solutions is required. Many different types of mapping heuristics have been developed in recent years. This paper investigates the task-scheduling problem as a multiobjective problem and its solution based on the Genetic Algorithm (GA) heuristics.GA has been successfully used in solving many of such multiple objective optimization problems in literature. We have used the inherent parallel nature of GA in developing a parallel genetic algorithm on a multicore processor to solve this optimization problem. Simulation results show that the parallel GA helps in generating optimized schedules at a faster convergence rate.