面向应用服务的虚拟机性能评估

详细信息    查看全文 | 推荐本文 |

英文篇名：VMs performance analysis based on application service 作者：曾文琦 ; 叶家炜 ; 杨阳 ; 刘发章 ; 吕智慧 英文作者：ZENG Wen-qi;YE Jia-wei;YANG Yang;LIU Fa-zhang;L Zhi-hui;School of Computer Science,Fudan University;Institute of Electronic Payment,China Unionpay; 关键词：云计算 ; 虚拟化 ; 虚拟机监控 ; 性能特征 ; 基础设施即服务 英文关键词：cloud computing;;virtualization;;virtual machine monitoring;;performance characteristics;;IaaS 中文刊名：SJSJ 英文刊名：Computer Engineering and Design 机构：复旦大学计算机科学技术学院;中国银联电子支付研究院; 出版日期：2014-10-16 出版单位：计算机工程与设计 年：2014 期：v.35;No.334 基金：国家云计算示范工程基金项目(C73623989020220110006) 语种：中文; 页：SJSJ201410054 页数：8 CN：10 ISSN：11-1775/TP 分类号：299-306

摘要

为了获取在云平台下提供应用服务的虚拟机所表征出的特定负载特性(这种负载特性可以用来标识此应用服务的行为),设计对现有虚拟机负载信息进行采集,达到了对虚拟机状态信息和负载信息有效获取;通过获取到的虚拟机系统信息,提出基于指定类型优化的性能评估算法,得出虚拟机的基本特征,对虚拟机性能负载进行量化。为虚拟机资源和物理机资源的充分使用提供解决思路。
        In cloud,lots of VMs are running to provide the application service.To get the VMs' specific load characteristics,which can be used to identify the behaviors of the application service,a collection method was designed to collect the VMs' load information and make it more effective to achieve the VMs' state information.A design of performance evaluation algorithm was presented based on the specific application service type.First,the VMs' system information was collected to calculate the VMs' basic characters,then the VMs' performance load was quantified.The method gives a new solution on how to use resources of VMs and PMs adequately.

引文

[1]Chaves D De SA,Uriarte RB,Westphall CB.Toward an architecture for monitoring private clouds[J].IEEE Communications Magazine December,2011,49(12):130-137.
    [2]Google App Engine.Google Inc[J/OL].http://code.google.com/appengine/,2013.
    [3]Knig B,Alcaraz Calero,Kirschnick JM,et al.Elastic monitoring framework for cloud infrastructures[J].IEEE Communications IET,2012,6(10):1306-1315.
    [4]Katsaros G,Kousiouris G,Gogouvitis S,et al.A service oriented monitoring framework for soft real-time applications[C]//Service Oriented Computing and Applciaitons,2010.
    [5]Napoleon D,Pavalakodi S.A new method for dimensionality reduction using k-means clustering algorithm for high dimensional data set[J].International Journal of Computer Applications(0975-8887),2011,13(7):41-46.
    [6]Selim S Z,Ismail M A.K-means-type algorithms:A generalized convergence theorem and characterization of local optimality[J].Pattern Analysis and Machine Inteligence,2009(1):81-87.
    [7]Prabhu P,Anbazhagan N.Improving the performance of kmeans clustering for high dimensional data set[J].International Journal on Computer Science and Engineering,2011,3(6):2321-2322.
    [8]Das S,Abraham A,Konar A.Automatic clustering using an improved differential evolution algorithm[J].IEEE Trans Syst,Man Cybern,A,Syst Humans,2008,38(1):218-237.
    [9]Matthews J N,Hu W,Hapuarachchi M,et al.Quantifying the performance isolation properties of virtualization systems[C]//Proceedings of the Workshop on Experimental Computer Science.New York,NY,USA:ACM,2007.
    [10]Mousa H,Doshi K,Ould-Ahmed-Vall E.Characterizing performance invirtualized execution[C]//2nd Workshop on Statistical and Machine Learning Approaches to Architectures and Compilation,2008.