摘要
针对分布式计算中的任务-处理器分配问题,提出了2种基于A*算法的改进算法.通过对A*算法原理的分析,得到将A*算法应用于任务分配问题的搜索树模型,和通过将搜索树在处理单元之间尽可能均匀地进行划分和通过避免不必要的节点扩展来得到最优解;实验结果表明,提出的2种改进算法,相比于传统的A*算法,不仅能够通过减少搜索空间节省内存和减少任务执行时间,而且能够通过并行分配来提高算法执行时的加速性.
Aiming at the problem for task-processor assignment in distributed computing,two improved algorithms based on A* algorithm are proposed.By analyzing the principle of A* algorithm,the search tree model of applying A* algorithm to task assignment problem is obtained,and by dividing the search tree as evenly as possible among the processing units and by avoiding unnecessary nodes extension,the optimal solution is obtained. The experimental results show that,compared with the traditional A* algorithm,the proposed two improved algorithms can not only save memory and task execution time by reducing search space,but also improve the speedup performance of algorithm execution by parallel assignment.
作者
高卫斌
柳晓龙
GAO Weibin;LIU Xiaolong(Department of Information Technology and Engineering,Ningde Vocational and Technical College,Ningde,Fujian,355000,China;School of Computer and Information,Fujian Agriculture and Forestry University,Fuzhou,Fujian,350002,China)
出处
《吉林化工学院学报》
CAS
2021年第1期52-57,73,共7页
Journal of Jilin Institute of Chemical Technology
基金
2017年宁德市科技项目“高校项目建设任务协同管理系统研究”(NSK2017137)。
关键词
分布式计算系统
任务分配
A*算法
最优解成本
加速性
distributed computing system
task assignment
A~*algorithm
optimal solution cost
speedup