摘要
研究了网格环境下任务调度问题,提出了一个任务调度机制:基于任务图将每一个可能的任务调度方案表示成一个任务-资源分配图,将网格任务调度问题转化为任务-资源分配图优化选取问题.提出了一种基于免疫遗传算法的、实现任务-资源分配图优化选取的任务调度算法.该算法将任务-资源分配图的最长路径作为抗原,每一个任务-资源分配图对应一个抗体.实验结果表明这个算法在全局优化能力及收敛速度上均有显著提高.
Studies the task scheduling in grid environment and proposes a task scheduling mechanism, i.e. each and every possible task scheduling scheme is expressed as a task-resource assignment graph, thus converting the task scheduling problem into a graphically optimal selection problem. Then, to find the optimal solution quickly and accurately, a task scheduling algorithm based on immune genetic algorithm is proposed to implement the optimal selection. This algorithm takes the longest path in the task-resource assignment graph as an antigen and every task-resource assignment as a corresponding antibody. Experimental results show that the approach proposed is effective in the capability of global optimization and significantly improves the convergence rate.
出处
《东北大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2007年第3期329-332,共4页
Journal of Northeastern University(Natural Science)
基金
国家"十五"科技攻关项目(2004BA721A05)
关键词
网格
任务调度
任务-资源分配图
优化选取
免疫遗传算法
grid
task scheduling
task-resource assignment graph
optimal selection
immune genetic algorithm
