吉林大学学报(理学版)

• 计算机科学 • 上一篇    下一篇

负载均衡的分布式系统任务调度优化算法

宋俊辉, 冯岩   

  1. 信阳师范学院 计算机与信息技术学院, 河南 信阳 464000
  • 收稿日期:2016-05-10 出版日期:2017-03-26 发布日期:2017-03-24
  • 通讯作者: 宋俊辉 E-mail:Songjunhui_88@sina.com

Task Scheduling Optimization Algorithm in Distributed System with Load Balancing

SONG Junhui, FENG Yan   

  1. College of Computer and Information Technology, Xinyang Normal University,Xinyang 464000, Henan Province, China
  • Received:2016-05-10 Online:2017-03-26 Published:2017-03-24
  • Contact: SONG Junhui E-mail:Songjunhui_88@sina.com

摘要: 针对当前分布式系统任务调度算法存在节点负载不平衡, 资源利用率低等缺陷, 提出一种基于负载均衡的分布式系统任务调度优化算法. 首先分析了当前分布式系统任务调度优化算法的研究现状; 然后对节点的实时性能指标进行估计, 并根据估计结果分配不同的任务; 最后通过仿真实验与其他算法进行对比. 实验结果表明, 相比于经典分布式系统任务调度算法, 该算法缩短了任务完成时间, 使得系统各节点的负载更均衡, 提高了节点资源的利用率.

关键词: 优化算法, 负载均衡, 分布式系统, 任务调度

Abstract: Aiming at the disadvantages of current task scheduling alg orithms in distributed system, such as node load imbalance and low resource util ization ratio, we proposed a novel task scheduling algorithm in distribu ted system based on load balancing. Firstly, the rese arch status of current task scheduling optimization algorithm in distributed sys tem was analyzed. Secondly, realtime performance of the node was estimated, a nd assigned different tasks according to the estimated results. Finally, the s imulation experiment was compared with other algorithms. Experimental res ults show that compared with the current classical task scheduling algor ithms in distributed system, the proposed algorithm reduces task completion ti me, makes load of each node more balanced, and improves the utilization ratio of node resource.

Key words: distributed system, optimization algorithm, task scheduling, load balancing

中图分类号: 

  • TP391