Scheduling Directed A-Cyclic Task Graphs on Heterogeneous Processors Using Task Duplication

Abstract

In distributed computing, the schedule by which tasks are assigned to processors is critical to minimizing the finish time of the application. However, the problem of discovering the schedule that gives the minimum finish time is NP-Complete. By combining several innovative techniques, such as insertion-based scheduling and multiple task duplication, this paper presents a new heuristic called the Heterogeneous N-predecessor Decisive Path (HNDP), for statically scheduling directed a-cyclic weighted task graphs (DAGs) on a set of heterogeneous processors to minimize makespan. We compare the performance of HNDP, under a range of input conditions, with two of the best existing heuristics namely HEFT and STDS. We show that HNDP outperforms the two heuristics in finish time and the number of processors employed over a wide range of parameters.