计算密集型与数据密集型混合网格作业调度算法

J4 ›› 2014, Vol. 36 ›› Issue (08): 1423-1429.

• 论文 • 下一篇

计算密集型与数据密集型混合网格作业调度算法

郝永生1，卢俊文2，刘冠峰3，温娜4

(1.南京信息工程大学网络信息中心,江苏南京 210044;2.厦门理工大学，福建厦门 361024；
3.苏州大学先进数据分析研究中心，江苏苏州 215006；4.南京信息工程大学大气科学学院，江苏南京 210044)

收稿日期:2013-03-20 修回日期:2013-05-20 出版日期:2014-08-25 发布日期:2014-08-25
基金资助:
福建省教育厅科技项目B类（JB09199）；国家自然科学基金资助项目（41005048）；科技部资助项目（GYHY201106037，GYHY200906023）

Three grid job scheduling methods heuristics
for data-intensive and computing-intensive jobs

HAO Yongsheng 1,LU Junwen2,LIU Guanfeng3,WEN Na4

(1.Network Center,Nanjing University of Information Science & Technology,Nanjing 210044；2.Xiamen University of Technology,Xiamen 361024;
3.Soochow Advanced Data Analytics Lab,Soochow University,Suzhou 215006;
4.College of Atmospherics Science,Nanjing University of Information Science & Technology,Nanjing 210044,China)

Received:2013-03-20 Revised:2013-05-20 Online:2014-08-25 Published:2014-08-25

摘要/Abstract

摘要：

针对计算密集型作业与数据密集型作业混合情况，在一个作业有时间限制的动态环境中，对传统的网格作业调度方法进行扩展，提出了三种网格作业调度启发式算法:Eminmin、Ebest、Esufferage。并在一个由多个Cluster组成的、通过高速网络连接的网格模型上，对三种算法进行验证。与Minmin算法的比较结果显示：三种算法均优于Minmin算法。与ASJS算法比较结果显示：Eminmin减少了等待时间与作业的makespan; Esufferage算法以减少作业完成量为代价，减少了作业的等待时间及makespan; Ebest在完成作业数量上与ASJS基本保持一致，但却增加了作业的等待时间与makespan。总体上，Eminmin具有比较大的优势。

关键词: 计算密集；数据密集；作业调度；平均执行时间

Abstract:

Most of the existing grid job scheduling methods focus on either dataintensive jobs or computingintensive jobs.In the dynamic environment where every job has its deadline,we extend the traditional grid job scheduling methods to propose three new grid job scheduling methods:Eminmin,Ebest and Esufferage.The three methods are validated on the Grid model with clusters that are connected by the high speed network.Simulation results demonstrate that our proposed methods are better than Minmin.The comparison between the three methods and ASJS shows that,Eminmin reduces the waiting time and makespan,Esufferage reduces the waiting time and the makespan greatly with the sacrifice of some jobs,and Ebest gives the same performance in unfinished jobs but has a larger value in waiting time and makespan than ASJS. In general,Eminmin has a better performance than Minmin and ASJS.

Key words: computing-intensive;data-intensive;job scheduling;average makespan

郝永生1，卢俊文2，刘冠峰3，温娜4. 计算密集型与数据密集型混合网格作业调度算法[J]. J4, 2014, 36(08): 1423-1429.

HAO Yongsheng 1,LU Junwen2,LIU Guanfeng3,WEN Na4. Three grid job scheduling methods heuristics
for data-intensive and computing-intensive jobs [J]. J4, 2014, 36(08): 1423-1429.

参考文献［12］

［1］	Maheswaran M,Ali S, Siegel H J, et al. Dynamic matching and scheduling of a class of independent tasks onto heterogeneous computing systems［C］∥Proc of the 8th IEEE Heterogeneous Computing Workshop,1999:3044.
［2］	Braun T D,Siegel H J,Beck N,et al. A comparison study of static mapping heuristics for a class of metatasks on heterogeneous computing systems［C］∥Proc of the 8th IEEE Heterogeneous Computing Workshop, 1999:1529.
［3］	Venugopal S, Buyya R. An SCPbased heuristic approach for scheduling distributed dataintensive applications on global grids［J］. Journal of Parallel and Distributed Computing,2008, 68(4):471487.
［4］	Kolodziej J, Xhafa F,Barolli L, et al. A taxonomy of data scheduling in data grids and data centers:Problems and intelligent resolution techniques［C］∥Proc of 2011 International Conference on Emerging Intelligent Data and Web Technologies (EIDWT),2011:6371.
［5］	Chang RuayShiung, Lin ChihYuan, Lin C F. An adaptive scoring job scheduling algorithm for grid computing［J］. Information Sciences,2012,207:7989,DOI:10.1016/j.ins.2012.04.019.
［6］	Valliyamma C,Somasundaram T S. A grid resource brokering strategy based on resource and network performance in grid［J］. Future Generation Computer Systems,2012, 28(3):491499.
［7］	Wang W, Luo J Z,Song A B. A GQoP guaranteed grid QoS adaptive scheduling algorithm［J］. Journal of Computer Research and Development, 2011,48(7):11681177.(in Chinese)
［8］	Hao Yongsheng,Liu Guanfeng,Wen Na. An enhanced load balancing mechanism based on deadline control on GridSim［J］.Future Generation Computer Systems,2012.28(4):657665,DOI:10.1016/j.future.2011.10.010.
［9］	Li X, Hu Z G, Hu Z J, et al. Grid workflow scheduling algorithms based on deadline Satisfaction［J］. Journal of Computer Research and Development, 2011,48(5):877884.(in Chinese)
［10］	GridSim Toolkit 5.0,GridSim:A toolkit for modeling and simulating grid computing environments ［EB/OL］.［20130501］. http://www.buyya.com/gridsim/.
［11］	Albodour R, James A, Yaacob N. High level QoSdriven model for grid applications in a simulated environment［J］. Future Generation Computer Systems,2012, 28(7):11331144.
［12］	Xiong Runqun, Luo Junzhou, Song Aibo, et al. QoS preferenceaware replica selection strategy in cloud computing［J］. Journal on Communications, 2011,32(7):93102.(in Chinese)
	附中文参考文献:
［7］	王巍,罗军舟,宋爱波.一种具有GQoP保证的网格QoS自适应调度算法［J］.计算机研究与发展,2011,48(7):11681177.
［9］	李玺,胡志刚,胡周君,等.基于截止时间满意度的网格工作流调度算法［J］.计算机研究与发展,2011,48(5):877884.
［12］	熊润群,罗军舟,宋爱波,等.云计算环境下QoS 偏好感知的副本选择策略［J］.通讯学报,2011,32(7):93102.

计算密集型与数据密集型混合网格作业调度算法

Three grid job scheduling methods heuristics
for data-intensive and computing-intensive jobs

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献［12］

相关文章 0

编辑推荐

Metrics

本文评价

计算密集型与数据密集型混合网格作业调度算法

Three grid job scheduling methods heuristics for data-intensive and computing-intensive jobs

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献 ［12］

相关文章 0

编辑推荐

Metrics

本文评价

Three grid job scheduling methods heuristics
for data-intensive and computing-intensive jobs

参考文献［12］