基于改进遗传算法的网格任务调度研究  

Grid Task Scheduling Based on Improved Genetic Algorithm

作　　者：叶春晓[1,2] 陆杰[1] 

机构地区：[1]重庆大学计算机学院,重庆400044 [2]重庆大学软件学院,重庆400044

出　　处：《计算机科学》2010年第7期233-235,共3页Computer Science

基　　金：国家"十一五"科技支撑计划项目(2006BAH02A20);国家自然科学基金(60803027)资助

摘　　要：网格任务调度是一个NP完全问题,它关注大规模的资源和任务调度,要求采用具有高效性的调度算法。提出了一种基于改进遗传算法的网格任务调度算法,在算法初始化种群产生时引入min-min算法和max-min算法,从而提高初始化种群的质量;算法迭代过程中采用了一种新的局部收敛判断以及改进的变异操作来防止局部收敛。仿真结果表明,该改进算法能更有效地解决网格任务调度问题。Grid task scheduling is a NP-complete problem which concerns the scheduling of tasks and resources in a large scale, and thus a scheduling algorithm of high efficiency is required. A grid task scheduling algorithm based on GA was proposed. In the process of population initialization, a new method which combines the min-min algorithm and the max-rain algorithm was addressed, and in the evolution of the population a new criterion predicting the premature convergence was presented and the corresponding improved mutation was designed to avoid premature convergence. The simulation results show that this improved algorithm can solve the problem of grid task scheduling more effectively.

关 键 词：网格 任务调度 遗传算法 局部收敛 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]