Genetic Algorithm for Task Scheduling on Distributed Heterogeneous Computing System

Distributed heterogeneous computing system are increasingly being employed for critical applications, such as aircraft control, industrial process control, and banking systems. Efficient application scheduling is a key issue for achieving high performance in this system. The problem is generally addressed in terms of task scheduling, where the tasks are the schedulable units of a program. The task scheduling problem has been extensively studied and a large number of scheduling heuristics have been presented in the literature. In this paper we propose a new task-scheduling algorithm namely, Genetic Algorithm for Task Scheduling (GATS) on heterogeneous computing system, which provides optimal results for applications represented by directed acyclic graph. The performance of the algorithm is illustrated by comparing the schedule length, speedup, and efficiency with existing algorithms such as CPOP, HEFT and PSGA. The comparison study based on randomly generated graphs and graphs of three real world applications such as Gaussian Elimination Algorithm, Fast Fourier Transformation, and Gauss Jordan algorithm shows that GATS algorithm substantially outperforms existing algorithms.