Task Scheduling for Heterogeneous Computing Based on Learning Classifier System

Task scheduling still remains one of the most challenging problems to achieve high performance in heterogeneous computing environments in spite of numerous efforts. This paper presents a novel scheduling algorithm based on learning classifier system for heterogeneous computing environment. In the presented algorithm, XCS classifier system is used to find the optimal task assignment on different processors, and the execution sequence of tasks on the same processor is set by the heuristic used in list scheduling approach. Empirical studies on benchmark task graphs show that the proposed algorithm is able to produce higher speedup compared with the related algorithms. Further experiments also indicate that the proposed algorithm maintains almost the same performance with different parameter settings.