《计算机应用研究》|Application Research of Computers

基于PEPA的云计算资源分配算法性能评价

PEPA model approach for performance evaluation of dynamic resource provision in cloud computing

免费全文下载 (已被下载 次)  
获取PDF全文
作者 王倩,石振国,孙万捷,刘赛男,王国涛
机构 南通大学 a.电子信息学院;b.计算机科学与技术学院,江苏 南通 226019
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2015)04-1179-05
DOI 10.3969/j.issn.1001-3695.2015.04.052
摘要 资源分配是云计算的核心之一,对云计算资源分配算法的性能进行评价可为云计算平台设计提供指导。讨论了两种云计算资源分配算法,提出了一种基于PEPA的资源分配算法的性能评价模型,该模型通过建立云计算系统中各组件之间的交互关系进行形式化分析和推理,获得了云计算系统性能的评价指标。实验通过分析资源分配过程中不同参数变化对系统性能的影响,结果表明,PEPA模型方法可以直接评估资源分配算法性能的优劣,并能够确定算法性能提升的关键因素,从而减少云平台设计过程的周期。
关键词 PEPA(性能评价进程代数);云计算;资源分配算法;性能评价
基金项目 国家自然科学基金资助项目(60975033)
江苏省高校自然科学基金资助项目(07KJB520096)
本文URL http://www.arocmag.com/article/01-2015-04-052.html
英文标题 PEPA model approach for performance evaluation of dynamic resource provision in cloud computing
作者英文名 WANG Qian, SHI Zhen-guo, SUN Wan-jie, LIU Sai-nan, WANG Guo-tao
机构英文名 a. School of Electronics & Information, b. School of Computer Science & Technology, Nontong University, Nantong Jiangsu 226019, China
英文摘要 One of the significant issues in cloud computing is the virtualized resources provision, since it will affect the performance of the entire system.Hence, performance evaluation of dynamic resources provision is necessary to provide some design guild lines.This paper adopted a method for analyzing the performance of resource provisioning process using PEPA approach.It analyzed two different PEPA models and then obtained the performance evaluation index by solving the underlying continuous time Markov chain(CTMC).Experimental results indicate that efficiency of resources provision algorithm determines the performance of the cloud computing system primarily, and the performance varies diversely under distinctive circumstances, which can be referenced for finding optimal strategy in cloud computing platform designing and deploying.
英文关键词 PEPA; cloud computing; resource provision algorithm; performance evaluation
参考文献 查看稿件参考文献
  [1] FOSTER I, ZHAO Yong, RAICU I, et al. Cloud computing and grid computing 360-degree compared[C] //Proc of Grid Computing Environments Workshop. Austin:GCE, 2008.
[2] 林闯, 李雅娟, 王忠民. 性能评价形式化方法的现状和发展[J] . 软件学报, 2002, 30(12):1919-1921.
[3] HILLSTON J, KLOUL L. Formal techniques for performance analysis:blending SAN and PEPA[J] . Formal Aspects of Computing, 2007, 19(1):3-33.
[4] GILMORE S, HILLSTON J. The PEPA workbench:a tool to support a process algebra-based approach to performance modelling[C] //Proc of the 7th Conference on Modelling Techniques and Tools for Computer Performance Evaluation. [S. l. ] :Springer-Verlag, 1994.
[5] 李春燕, 何一舟, 戴彬. Hadoop平台的多队列作业调度优化方法[J] . 计算机应用研究, 2014, 31(3):705-738.
[6] WANG Guo-hui, NG T S E. The impact of virtualization on network performance of Amazon EC2 data center[C] //Proc of the 29th Conference on Information Communications. [S. l. ] :IEEE Press, 2010.
[7] ASSUNCAO M D D, COSTANZO A D, BUYYA R. Evaluating the cost-benefit of using cloud computing to extend the capacity of clusters[C] //Proc of the 18th ACM International Symposium on High Performance Distributed Computing. New York:ACM Press, 2009.
[8] LOSUP A, OSTERMANN S, YIGITBASI N, et al. Performance analysis of cloud computing services for many-tasks scientific computing[J] . IEEE Trans on Parallel and Distributed Systems, 2011, 22(6):931-945.
[9] LI Ang, YANG Xiao-wei, KANDULA S, et al. CloudCmp:comparing public cloud providers[C] //Proc of the 10th ACM SIGCOMM Conference on Internet Measurement. New York:ACM Press, 2010.
[10] LI Zhi-chun, ZHANG Ming, ZHU Zhao-sheng, et al. WebProphet:automating performance prediction for Web services[C] //Proc of the 7th USENIX Conference on Networked Systems Design and Implementation. Ber:USENIX Association, 2010.
[11] STANTCHEV V. Performance evaluation of cloud computing offerings[C] //Proc of the 3rd International Conference on Advanced Engineering Computing and Applications in Sciences. 2009.
[12] KHAZAEI H, MISIC J, MISIC V B. Performance analysis of cloud computing centers using M/G/m/m+r queuing systems[J] . IEEE Trans on Parallel and Distributed Systems, 2012, 23(5):936-943.
[13] WARNEKE D, KAO O. Exploiting dynamic resource allocation for efficient parallel data processing in the cloud[J] . IEEE Trans on Parallel and Distributed Systems, 2011, 22(6):985-997.
[14] BUYYA R, GARG S K, CALLHEIROS R N, et al. SLA-oriented resource provisioning for cloud computing:challenges, architecture, and solutions[C] //Proc of International Conference on Cloud and Service Computing. 2011.
[15] GARG S K, VERSTEEG S, BUYYA R. SMICloud:a framework for comparing and ranking cloud services[C] //Proc of the 4th IEEE International Conference on Utility and Cloud Computing. 2011:210-218.
[16] LOSUP A, YIGITBASI N, EPEMA D. On the performance variability of production cloud services[C] // Proc of the 11th International Conference on Cluster, Cloud and Grid Computing. 2011:104-113.
[17] SCHAD J, DITTRICH J, QUIANERUIZ J A. Runtime measurements in the cloud:observing, analyzing, and reducing variance[J] . Proc VLDB Endow, 2010, 3(1-2):460-471.
[18] GARG S K, BUYYA R. NetworkCloudSim:modelling parallel applications in cloud simulations[C] //Proc of the 4th IEEE International Conference on Utility and Cloud Computing. 2011:105-113.
[19] NURMI D, WOLSKI R, GRZEGORCZYK C, et al. The eucalyptus open-source cloud-computing system[C] // Proc of Grid Computing Environments Workshop. 2009.
[20] KYI H M, NAING T T. Stochastic Markov model approach for efficient virtual machines scheduling on private cloud[J] . International Journal on Cloud Computing:Services and Architecture, 2011, 1(3):1-13.
[21] DONATELLI S, RIBAUDO M, HILLSTON J. A comparison of performance evaluation process algebra and generalized stochastic Petri nets[C] //Proc of the 6th International Workshop on Petri Nets and Performance Models. 1995.
[22] 郭辉. 进程代数及其在性能中的应用综述[J] . 微计算机应用, 2007, 28(9):901-905.
[23] 马梁, 李明, 宋洁, 等. 进程代数在性能评价中的应用研究[J] . 河北工业大学学报, 2006, 35(4):35-39.
收稿日期 2014/3/2
修回日期 2014/4/23
页码 1179-1183,1187
中图分类号 TP393.07
文献标志码 A