Energy-Aware Task Migration Through Ant-Colony Optimization for Multiprocessors

Abstract

In this work, we introduce a novel strategy to improve the power dissipation of the Multiprocessor System on Chips (MPSoC) through a modified Ant-Colony Optimization (ACO) for task migration. Combined with a First-Fit task allocation heuristic, the ACO algorithm tries to split tasks and migrate them to processors with low task utilization to minimize the overall power consumption of the MPSoC. Finally, the task set, including split tasks, is scheduled through an Early-Deadline-First (EDF) scheduler. This paper describes the implementation and verification of the proposed work, and the results of the experiment attest to the improvements gained over the traditional allocating and scheduling algorithms in the literature.