基于蚂蚁算法的网格作业调度研究  

Research on Job Scheduling Based on Ant Algorithm in Computational Grid

作　　者：赵晨阳[1] 杨裔[1] 刘礼[1] 李廉[1] 

机构地区：[1]兰州大学信息学院,甘肃兰州730000

出　　处：《微电子学与计算机》2007年第9期78-81,共4页Microelectronics & Computer

基　　金：国家自然科学基金项目(90612016);国家科技基础条件平台项目(60473095)

摘　　要：网格环境下的作业调度是一个NP难问题,蚂蚁算法内在的并行性和可扩充性使其非常适合网格作业调度。将蚂蚁算法应用于网格环境作业调度,提出一种通过作业代理的移动进行网格作业调度的方案,该蚂蚁算法不仅在分配网格计算资源时进行信息素的局部更新,还在网格计算资源完成作业后进行信息素的整体更新。通过模拟实验测试和选取蚂蚁算法的各种影响参数,取得了比较理想的实验结果。实验证明该算法能够有效地实现作业的合理调度和网格系统的负载平衡。Job scheduling of Grid is a NP-hard problem. The inherent parallelism and scalability of ant algorithm make the algorithm very suitable for job scheduling in the Computational Grid. In this paper, we apply the ant algorithm to solve the job scheduling problem in Grid and bring forward a job scheduling implement through the movement of the job-agents.The algorithm updates local pheromone when the jobs are assigned to the grid resources, and updates global pheromone after these jobs are executed. By simulation experiment, we tested and selected every parameter of ant algorithm, and got fair results. Actually, the algorithm makes the job dispatcher and load balancing effectivelv.

关 键 词：网格计算 作业调度 蚂蚁算法 负载平衡 

分 类 号：TP393.1[自动化与计算机技术—计算机应用技术]