计算机应用 ›› 2015, Vol. 35 ›› Issue (8): 2153-2157.DOI: 10.11772/j.issn.1001-9081.2015.08.2153

• 先进计算 • 上一篇    下一篇

计算集群中一种基于任务运行时间的组合预测方案

余莹1, 李肯立2, 徐雨明2   

  1. 1. 衡阳师范学院 计算机科学系, 湖南 衡阳 421002;
    2. 湖南大学 信息科学与工程学院, 长沙 410082
  • 收稿日期:2015-02-10 修回日期:2015-04-05 出版日期:2015-08-10 发布日期:2015-08-14
  • 通讯作者: 余莹(1982-),女,湖南益阳人,讲师,硕士,主要研究方向:任务调度、并行计算,yuying_kszx@126.com
  • 作者简介:李肯立(1971-),男,湖南娄底人,教授,博士生导师,博士,主要研究方向:并行处理、网格计算、DNA计算、实时与混成嵌入式系统; 徐雨明(1966-),男,浙江衢州人,副教授,博士研究生,主要研究方向:任务调度。
  • 基金资助:

    国家自然科学基金资助项目(61370095,61370098,61070057,90715029);湖南省教育厅科学研究项目(13C074);衡阳市科技发展计划项目(2011KJ22);湖南省教育科学"十二五"规划课题(XJK014CGD006)。

Combined prediction scheme for runtime of tasks in computing cluster

YU Ying1, LI Kenli2, XU Yuming2   

  1. 1. Department of Computer Science, Hengyang Normal University, Hengyang Hunan 421002, China;
    2. College of Computer Science and Electronic Engineering, Hunan University, Changsha Hunan 410082, China
  • Received:2015-02-10 Revised:2015-04-05 Online:2015-08-10 Published:2015-08-14

摘要:

针对现有单一预测策略不适用于所有异构任务的问题,提出一种基于本地任务与远程任务运行时间的组合预测方案(CPS)和预测精度保证(PAA)的概念。使用GridSim工具集来实现CPS,将PAA作为定量评价由某一特定预测策略提供的预测运行时间精度的标准。仿真实验表明:与本地任务预测策略如Last和滑动窗口中值(SM)相比,CPS的平均相对残差下降了1.58%、1.62%;与远程任务预测策略如平均运行时间(RM)和加权移动平均值(ES)相比,CPS的平均相对残差下降了1.02%、2.9%。因此,PAA能从综合策略所提供的结果中选择接近最优值的预测,CPS增强了计算环境中本地任务和远程任务运行时间的PAA。

关键词: 计算集群, 组合预测方案, 预测精度保证, 任务, 运行时间

Abstract:

A Combined Prediction Scheme (CPS) and a concept of Prediction Accuracy Assurance (PAA) were put forward for the runtime of local and remote tasks, on the issue of inapplicability of the singleness policy to all the heterogeneous tasks. The toolkit of GridSim was used to implement the CPS, and PAA was a quantitative evaluation standard of the prediction runtime provided by a specific strategy. The simulation experiments showed that, compared with the local task prediction strategy such as Last and Sliding Median (SM), the average relative residual error of CPS respectively reduced by 1.58% and 1.62%; and compared with the remote task prediction strategy such as Running Mean (RM) and Exponential Smoothing (ES), the average relative residual error of CPS respectively reduced by 1.02% and 2.9%. The results indicate that PAA can select the near-optimal value from the results of comprehensive prediction strategy, and CPS enhances the PAA of the runtime of local and remote tasks in the computing environments.

Key words: computing cluster, Combined Prediction Scheme (CPS), Prediction Accuracy Assurance (PAA), task, runtime

中图分类号: