基于改进的遗传算法的任务分配与调度  

Tasks Matching and Scheduling of Distributed System Using a Modified Genetic Algorithm

作　　者：刘晶晶[1] 翟正军[1] 

机构地区：[1]西北工业大学计算机学院,陕西西安710072

出　　处：《微电子学与计算机》2006年第6期216-219,共4页Microelectronics & Computer

摘　　要：任务分配与调度是分布式系统中的关键问题,应用遗传算法来解决这一问题是近年来出现的新思路。文章介绍了一种改进的遗传算法,该算法在扩大选择范围的同时,将适应度引入到对父代之间的交叉操作中,克服了标准遗传算法(SGA)交叉操作中较大的盲目性。实验结果表明,改进的遗传算法在精度和收敛速度上都较SGA有较大程度的提高。Task matching and scheduling is a key technology of distributed system, and the problem is being resolved with genetic algorithms these years. In this paper, a modified genetic algorithm is introduced. The algorithm expands the selection of chromosomes, and considers the fitness of parent chromosomes. The experiment has proved that the solving quality and efficiency are improved markedly with the modified genetic algorithm in comparison with the SGA.

关 键 词：分布式系统 任务分配与调度 遗传算法 适应度 

分 类 号：TP338.8[自动化与计算机技术—计算机系统结构]