Abstract: Task scheduling is an important part in Grid computing.As it has been proven to be NP- complete, it is adopt to be solved by heuristics.For heterogeneous and distributed grid,a double-deck genetic structure based grid task scheduling algorithm was presented with the principle of genetic algorithm and the method of ranking according to height-value. Simulation results show that the proposed algorithm performs better than other scheduling algorithm in heterogenous grid.

Key words: grid computing, task scheduling, double-deck genetic structure, genetic algorithm

摘要： 在网格计算中，任务调度是一个重要的组成部分，并被证明为NP完全问题，以启发式方法求解较为适合。针对网格环境异构、分布等特点，本文运用遗传算法原理，结合高度分层排序，提出了一种基于双层进化结构的网格任务调度算法。模拟实验结果表明，在网格异构环境下，本算法优于其它几种算法。

关键词: 网格计算, 任务调度, 双层进化结构, 遗传算法