A Novel Task Prioritizing Algorithm Based on Critical Path for a List Scheduling

초록

In a distributed heterogeneous computing system, performance of a parallel application significantly depends on its task scheduling algorithm. Therefore, in order to improve it, it is essential to consider some factors that can have effect on the performance of the parallel application in a given environment. One of the most important factors that affects the total execution time is a critical path. In this paper, we propose a Critical node Look-ahead Task Prioritizing (CLTP) algorithm for task prioritizing in a list scheduling. The proposed algorithm improves overall performance by applying leveling method to improve parallelism of task execution and by reducing the delay caused by waiting for execution of critical nodes. To evaluate the performance of the CLTP, we compared the CLTP with a Dynamic Critical Path Duplication (DCPD) and a Heterogeneous Critical Parents with Fast Duplicator (HCPFD) with respect to the average schedule length ratio in our simulation. The results of the simulations show that the CLTP is better than the DCPD by 11.40% and the HCPFD by 15.25%.