网格计算环境下资源管理的研究
|
摘要
资源管理是网格计算中最重要的组成部分之一。网格资源管理系统的有效性和适用性主要取决于其采用的资源调度策略。但是网格资源的分布性、异构性、自治性及动态性特点,决定了网格资源调度的复杂性。因此,本文引入一种基于计算市场的经济学模型来解决网格资源的管理和调度问题。
本文首先介绍课题的背景、网格的概念、特点、发展阶段、国内外的研究现状以及网格计算和云计算之间的区别与联系。
其次对五层沙漏结构和开放网格服务体系结构进行详细的论述。
第三,阐述网格资源管理的相关概念和技术、目前主要的三种网格资源管理体系结构模型以及典型的网格资源管理系统。
第四,探讨引入经济学进行网格资源管理的原因,基于经济学构建的Nimrod-G网格资源管理系统,对代价最优、时间最优和代价时间最优这三种期限与预算限制算法进行详细的研究和模拟仿真,并对仿真结果进行了分析和比较。
第五,为了解决资源调度过程中,资源消费者与生产者之间的信任问题,我们应用资源信誉度的概念和评估方法,结合当前较成熟的期限与预算限制算法,实现了基于信誉度的期限与预算限制算法。
第六,在剖析使用模拟器进行网格仿真实验的原因、GridSim的特征、体系结构后,对GridSim的源代码和使用方法进行了分析,并结合NetBeans IDE、iBATIS数据持久化框架、MySQL数据库,对其代码进行相关的修改补充,以提供网格模拟实验的GUI、关于资源信誉度的获取、注册、更新和注销等操作。在此基础上对基于信誉度的DBC算法进行仿真模拟,并对仿真结果进行分析。实验结果表明,把信誉度融入经济模型中是可行的,它增强了网格资源消费者与生产者之间的信任,提高了调度性能和资源利用率。
最后,对全文进行总结,并提出今后进一步研究的方向。
Resource management is one of the most important parts of grid computing. The validity and applicability of the grid resource management system depend on its scheduling strategy mostly. But the characteristics of grid such as high heterogeneity, adaptability, distribution and dynamic decided the complexity of grid resource scheduling. Therefore, the model of computer economic is used for resolving resource management and scheduling.
Firstly, in this paper we introduce the concept, the characteristics, three development phases of grid, research at home and abroad, differences and relationship between grid computing and cloud computing.
Secondly, we give a detailed account of Five-Level Sandglass Architecture and Open Grid Services Architecture.
Thirdly, we introduce the relevant concepts of grid resource management, the three major grid resource management architecture and the typical grid resource management systems.
Fourthly, we expound why introduce economics for grid resource management and introduce the economics based grid resource management system Nimrod-G, do simulated experiments on cost optimization, time optimization, cost time optimization, these three different deadlines and budget constraints (DBC) algorithm, and analyze experimental results.
Fifthly, we apply the theory and the evaluation method of credit, combine with mature DBC algorithm, bring forward the DBC algorithm based on the credit, and implement the algorithm in order to solve the trust issue between resource consumers and producers in the resource scheduling process and to remedy the deficiency of the original algorithm.
Sixthly, we introduce the reasons of using grid simulator for the experiment, the characteristics, architecture and the direction of use on the grid simulator GridSim. After the analysis the source code of the GridSim, combine with NetBeans IDE, data persistence framework iBATIS, MySQL database, we improve the source code of GridSim and provide additional function to support the GUI of grid simulation experiment and the operation of credit. Then we make simulation experiments on the DBC algorithm based on credit. Experimental results show that putting credit into the economic model is feasible, it increases the confidence between the grid resource consumers and producers, and improve the scheduling performance and resource utilization.
Lastly, we conclude the paper and prospect a few directions for future research.
本文首先介绍课题的背景、网格的概念、特点、发展阶段、国内外的研究现状以及网格计算和云计算之间的区别与联系。
其次对五层沙漏结构和开放网格服务体系结构进行详细的论述。
第三,阐述网格资源管理的相关概念和技术、目前主要的三种网格资源管理体系结构模型以及典型的网格资源管理系统。
第四,探讨引入经济学进行网格资源管理的原因,基于经济学构建的Nimrod-G网格资源管理系统,对代价最优、时间最优和代价时间最优这三种期限与预算限制算法进行详细的研究和模拟仿真,并对仿真结果进行了分析和比较。
第五,为了解决资源调度过程中,资源消费者与生产者之间的信任问题,我们应用资源信誉度的概念和评估方法,结合当前较成熟的期限与预算限制算法,实现了基于信誉度的期限与预算限制算法。
第六,在剖析使用模拟器进行网格仿真实验的原因、GridSim的特征、体系结构后,对GridSim的源代码和使用方法进行了分析,并结合NetBeans IDE、iBATIS数据持久化框架、MySQL数据库,对其代码进行相关的修改补充,以提供网格模拟实验的GUI、关于资源信誉度的获取、注册、更新和注销等操作。在此基础上对基于信誉度的DBC算法进行仿真模拟,并对仿真结果进行分析。实验结果表明,把信誉度融入经济模型中是可行的,它增强了网格资源消费者与生产者之间的信任,提高了调度性能和资源利用率。
最后,对全文进行总结,并提出今后进一步研究的方向。
Resource management is one of the most important parts of grid computing. The validity and applicability of the grid resource management system depend on its scheduling strategy mostly. But the characteristics of grid such as high heterogeneity, adaptability, distribution and dynamic decided the complexity of grid resource scheduling. Therefore, the model of computer economic is used for resolving resource management and scheduling.
Firstly, in this paper we introduce the concept, the characteristics, three development phases of grid, research at home and abroad, differences and relationship between grid computing and cloud computing.
Secondly, we give a detailed account of Five-Level Sandglass Architecture and Open Grid Services Architecture.
Thirdly, we introduce the relevant concepts of grid resource management, the three major grid resource management architecture and the typical grid resource management systems.
Fourthly, we expound why introduce economics for grid resource management and introduce the economics based grid resource management system Nimrod-G, do simulated experiments on cost optimization, time optimization, cost time optimization, these three different deadlines and budget constraints (DBC) algorithm, and analyze experimental results.
Fifthly, we apply the theory and the evaluation method of credit, combine with mature DBC algorithm, bring forward the DBC algorithm based on the credit, and implement the algorithm in order to solve the trust issue between resource consumers and producers in the resource scheduling process and to remedy the deficiency of the original algorithm.
Sixthly, we introduce the reasons of using grid simulator for the experiment, the characteristics, architecture and the direction of use on the grid simulator GridSim. After the analysis the source code of the GridSim, combine with NetBeans IDE, data persistence framework iBATIS, MySQL database, we improve the source code of GridSim and provide additional function to support the GUI of grid simulation experiment and the operation of credit. Then we make simulation experiments on the DBC algorithm based on credit. Experimental results show that putting credit into the economic model is feasible, it increases the confidence between the grid resource consumers and producers, and improve the scheduling performance and resource utilization.
Lastly, we conclude the paper and prospect a few directions for future research.
引文
[1]L.Smarr and C.Catlett,"MetaComputing"[J],Communications of the ACM,35(6),pp.44-52,1992
[2]I.Foster,C.Kesselman(editor).The Grid:Blueprint for a New Computing Infrastructure[M].Morgan Kaufmann,San Francisco,CA,1999
[3]I.Foster,C.Kesselman,J.Nick,S.Tuecke,The Physiology of the Grid:An Open Grid Services Architecture for Distributed Systems Integration[Z],January,2002
[4]I.Foster,C.Kesselman,S.Tuecke,The Anatomy of the Grid:Enabling Scalable Virtual Organizations[J],International J.Supercomputer Applications,15(3),2001,200-222
[5]I.Foster.Internet Computing and the Emerging Grid[D].Nature Web Matters,December 7,2000.
[5]I.Foster.What is the Grid? A Three Point Checklist[D].GRID Today,July 20,2002
[7]David De Roure,Mark A.Baker,Nicholas R.Jennings,Nigel R.Shadbolt.The Evolution of the Grid[Z]
[8]The Globus Alliance[D].http://www.globus.org/
[9]K Czajkowski,I Foster,N Karonis et al.A Resource Management Architecture for Metacomputing System[J].In:Proceeding of IPPS/SPSP98 Workshop on Job Scheduling Strategies for Parallel Processing,Orlando,Florida,USA,1998
[10]http://legion.virginia.edu/
[11]I.Foster,C.Kesselman-The Globus Project:A Status Report[J].In:Proceeding of IPPS/SPSP98Heterogenous Computing Workshop,Orlando,Florida,USA,1998
[12]R Buyya,David Abramson,Jonathan Giddy.Nimrod-G:An Architecture for a Resource Management and Scheduling System in a Global Computational Grid[J].In:The 4~(th) International Conference on High Performance Computing in Asia-Pacific Region(HPC Asia 2000) Beijing,China IEEE Computing Society Press,USA,2000
[13]L Boloni and D C Marinescu.An Object-oriented Framework for Building Collaborative Network Agents[J].In:A Kandel,K Homann,D Miynek,and N H Teodorescu,editors.Intelligent Systems and Interfaces,Kluwer Publishing 2000.31-64
[14]http://www.cs.wisc.edu/condor
[15]http://www.sbsm.gov.cn/article/zszygx/zjlt/200812/20081200045771.shtml
[16]我国网格研究现状[Z],刘鹏。http://www.chinagrid.net/grid/paperppt/OurCountry.pdf
[17]K.Czajkowski,A.Dan,J.Rofrano,S.Tuecke,M.Xu.Agreement-based Grid Service Management (OGSI-Agreement)[J],June 12,2003.
[18]郝泳涛,方丁,林琳。基于OGSI的网格计算研究综述[J]。计算机应用研究,2005,18(04):18-20
[19]Karl Czajkowski,Donald F.Fergusion,Inn Foster,et al.WS-Resource Framework[Z],http://www.globus.org/wsrf/
[20]Rajkumar Buyya.Economic-Based Distributed Resource Management and Scheduling for Grid Computing[J].http://www.buyya.com/
[21]Manzur Murshed and Rajkumar Buyya,Using the GridSim Toolkit for Enabling Grid Computing Education,International Conference on Communication Networks and Distributed Systems Modeling and Simulation(CNDS 2002)[J],January 27-31,2002,San Antonio,Texas,USA.
[22]Rajkumar Buyya and Manzur Murshed,GridSim:A Toolkit for the Modeling and Simulation of Distributed Resource Management and Scheduling for Grid Computing[J],The Journal of Concurrency and Computation:Practice and Experience(CCPE),Volume 14,Issue 13-15,Wiley Press,Nov-Dec,2002
[23]T.Braun,H.Siegel,N.Beck,L.Boloni,M.Maheswaran,A.Reuther,J.Robertson,M.Theys,B.Yao,D.Hensgen,and R.Freund.A Comparison Study of Static Mapping Heuristics for a Class of Meta-tasks on Heterogeneous Computiag Systems[J].In 8th IEEE Heterogeneous Computing Workshop(HCW'99),pages 15-29,Apr.1999
[24]K.Krauter,R.Buyya,M.Maheswaran.A Taxonomy and Survey of Grid Resource Management Systems[J].Software-Practice and Experience,2002,32(2):135-164
[25]S.Chapin,D.Katramatos,J.Karpovich,A.Grimshaw.Resource Management in Legion[J].Future Generation Computer Systems,1999,15(5):583-594
[26]曹鸿强,基于多经济智能主体的计算网格资源管理[D]。中国人民解放军国防科学技术大学研究生院,2000年10月
[27]孙渝,李志平。网格资源管理体系结构模型研究[J]。计算机工程与应用,2003,17:26-29
[28]R Buyya,S Chapin D DiNucci.Architecture Models for Resource Management in the Grid[C].In:The First IEEE/ACM International Workshop on Grid Computing(GRID 2000),Springer Verlag LNCS Series,Germany,Bangalore,India 2000
[29]施运梅,刘梅彦,杨根兴。网格资源管理模型研究[J]。微计算机信息,2006,22:109-111,122
[30]The DOE Common Component Architecture Project[Z].http://www.extreme.indiana.edu/~gannon /cca_report.html
[31]J.Basney and M Livny.Deploying a High Throughput Computing Cluster[J].High Performance Cluster Computing,R Buyya(editor).Vol.1,Chapters 5,Prentice Hall PTR,May,1999
[32]M Litzkow,M Livny,M Mutka.Condor-A Hunter of Idle Workstations[C].Proceeding of the 8~(th)International Conference of Distributed Computing Systems(ICDCS1988),January 1988,San Jose,CA,IEEE CS Press,USA,1988
[33]F.Berman and R.Wolski,The AppLes Project:A Status Report[C],Proceedings of the 8~(th) NEC Research Symposium,Berlin,Germany,May 1997
[34]M.Neary,A.Phipps,S.Richman,P.Cappello,Javelin 2.0:Java-Based Parallel Computing on the Internet[C],Proceedings of European Parallel Computing Conference(Euro-Par 2000),Germany,2000
[35]W.Hoschek,J.Jaen-Martinez,A.Samar,H.Stockinger,and K.Stockinger,Data Management in an International Data Grid Project[C],Proceedings of the First IEEE/ACM international Workshop on Grid Computing,(Springer Verlag Press,Germany),India,2000
[36]R.Buyya,D.Abramson,and J.Giddy,An Economy Driven Resource Management Architecture for Global Computational Power Grids[C],Proceedings of the 2000 International Conference on Parallel and Distributed Processing Techniques and Applications(PDPTA 2000),June 26-29,2000,Las Vegas,USA,CSREA Press,USA,2000
[37]D.Ferguson,C.Nikolaou,J.Sairamesh,and Y.Yemini,Economic Models for Allocating Resources in Computer Systems[C],In Market-based Control:a Paradigm for Distributed Resource Allocation,World Scientific Press,Singapore,1996
[38]D Abramson,I.Foster,J.Giddy,A.Lewis,R.Sosic,R.Sutherst,N.White,The Nimrod Computational Workbench:A Case Study in Desktop Metacomputing[J],Australian Computer Science Conference(ACSC 97),Macquarie University,Sydney,Feb 1997
[39]D.Abramson,R.Sosic,J.Giddy and B.Hall,Nimrod:A Tool for Performing Parameterized Simulations Using Distributed Workstations[C],Proceedings of the 4~(th) IEEE International Symposium on High Performance Distributed Computing,Virginia,August 1995,IEEE CS Press, USA,1995
[40]D.Abramson,P.Roe,L.Kotler,and D.Mather,ActiveSheets:Super-Computing with Spreadsheets[C].2001 High Performance Computing Symposium(HPC'01),Advanced Simulation Technologies Conference,April 2001
[41]G.Zacharia,P.Maes.Trust Management Through Reputation Mechanisms[J].Applied Artificial Intelligence Journal,2000,14(9):881-908
[42]B.Padovan,S.Sackmann,et al.A Prototype for an Agent-based Secure Electronic Marketplace including Reputation Tracking Mechanisms[J].International Journal of Electronic Commerce,2002,6(4):93
[43]马满福,吴健,陈丁剑,邓磊,胡正国。网格经济模型中基于信誉度的资源选择[J]。计算机工程,2006,32(17):175-177
[44]马满福,吴健,胡正国,陈丁剑,蒋芸。网格计算经济中的信誉度体系结构[J]。计算机应用,2005,25(4),940-943
[45]F.Azzedin,M.Maheswaran.Towards Trust-aware Resource Management in Grid Computing Systems[C].Cluster Computing and the Grid 2~(nd) IEEE/ACM International Symposium CC-Grid2002,May 21-24 2002,419-424
[2]I.Foster,C.Kesselman(editor).The Grid:Blueprint for a New Computing Infrastructure[M].Morgan Kaufmann,San Francisco,CA,1999
[3]I.Foster,C.Kesselman,J.Nick,S.Tuecke,The Physiology of the Grid:An Open Grid Services Architecture for Distributed Systems Integration[Z],January,2002
[4]I.Foster,C.Kesselman,S.Tuecke,The Anatomy of the Grid:Enabling Scalable Virtual Organizations[J],International J.Supercomputer Applications,15(3),2001,200-222
[5]I.Foster.Internet Computing and the Emerging Grid[D].Nature Web Matters,December 7,2000.
[5]I.Foster.What is the Grid? A Three Point Checklist[D].GRID Today,July 20,2002
[7]David De Roure,Mark A.Baker,Nicholas R.Jennings,Nigel R.Shadbolt.The Evolution of the Grid[Z]
[8]The Globus Alliance[D].http://www.globus.org/
[9]K Czajkowski,I Foster,N Karonis et al.A Resource Management Architecture for Metacomputing System[J].In:Proceeding of IPPS/SPSP98 Workshop on Job Scheduling Strategies for Parallel Processing,Orlando,Florida,USA,1998
[10]http://legion.virginia.edu/
[11]I.Foster,C.Kesselman-The Globus Project:A Status Report[J].In:Proceeding of IPPS/SPSP98Heterogenous Computing Workshop,Orlando,Florida,USA,1998
[12]R Buyya,David Abramson,Jonathan Giddy.Nimrod-G:An Architecture for a Resource Management and Scheduling System in a Global Computational Grid[J].In:The 4~(th) International Conference on High Performance Computing in Asia-Pacific Region(HPC Asia 2000) Beijing,China IEEE Computing Society Press,USA,2000
[13]L Boloni and D C Marinescu.An Object-oriented Framework for Building Collaborative Network Agents[J].In:A Kandel,K Homann,D Miynek,and N H Teodorescu,editors.Intelligent Systems and Interfaces,Kluwer Publishing 2000.31-64
[14]http://www.cs.wisc.edu/condor
[15]http://www.sbsm.gov.cn/article/zszygx/zjlt/200812/20081200045771.shtml
[16]我国网格研究现状[Z],刘鹏。http://www.chinagrid.net/grid/paperppt/OurCountry.pdf
[17]K.Czajkowski,A.Dan,J.Rofrano,S.Tuecke,M.Xu.Agreement-based Grid Service Management (OGSI-Agreement)[J],June 12,2003.
[18]郝泳涛,方丁,林琳。基于OGSI的网格计算研究综述[J]。计算机应用研究,2005,18(04):18-20
[19]Karl Czajkowski,Donald F.Fergusion,Inn Foster,et al.WS-Resource Framework[Z],http://www.globus.org/wsrf/
[20]Rajkumar Buyya.Economic-Based Distributed Resource Management and Scheduling for Grid Computing[J].http://www.buyya.com/
[21]Manzur Murshed and Rajkumar Buyya,Using the GridSim Toolkit for Enabling Grid Computing Education,International Conference on Communication Networks and Distributed Systems Modeling and Simulation(CNDS 2002)[J],January 27-31,2002,San Antonio,Texas,USA.
[22]Rajkumar Buyya and Manzur Murshed,GridSim:A Toolkit for the Modeling and Simulation of Distributed Resource Management and Scheduling for Grid Computing[J],The Journal of Concurrency and Computation:Practice and Experience(CCPE),Volume 14,Issue 13-15,Wiley Press,Nov-Dec,2002
[23]T.Braun,H.Siegel,N.Beck,L.Boloni,M.Maheswaran,A.Reuther,J.Robertson,M.Theys,B.Yao,D.Hensgen,and R.Freund.A Comparison Study of Static Mapping Heuristics for a Class of Meta-tasks on Heterogeneous Computiag Systems[J].In 8th IEEE Heterogeneous Computing Workshop(HCW'99),pages 15-29,Apr.1999
[24]K.Krauter,R.Buyya,M.Maheswaran.A Taxonomy and Survey of Grid Resource Management Systems[J].Software-Practice and Experience,2002,32(2):135-164
[25]S.Chapin,D.Katramatos,J.Karpovich,A.Grimshaw.Resource Management in Legion[J].Future Generation Computer Systems,1999,15(5):583-594
[26]曹鸿强,基于多经济智能主体的计算网格资源管理[D]。中国人民解放军国防科学技术大学研究生院,2000年10月
[27]孙渝,李志平。网格资源管理体系结构模型研究[J]。计算机工程与应用,2003,17:26-29
[28]R Buyya,S Chapin D DiNucci.Architecture Models for Resource Management in the Grid[C].In:The First IEEE/ACM International Workshop on Grid Computing(GRID 2000),Springer Verlag LNCS Series,Germany,Bangalore,India 2000
[29]施运梅,刘梅彦,杨根兴。网格资源管理模型研究[J]。微计算机信息,2006,22:109-111,122
[30]The DOE Common Component Architecture Project[Z].http://www.extreme.indiana.edu/~gannon /cca_report.html
[31]J.Basney and M Livny.Deploying a High Throughput Computing Cluster[J].High Performance Cluster Computing,R Buyya(editor).Vol.1,Chapters 5,Prentice Hall PTR,May,1999
[32]M Litzkow,M Livny,M Mutka.Condor-A Hunter of Idle Workstations[C].Proceeding of the 8~(th)International Conference of Distributed Computing Systems(ICDCS1988),January 1988,San Jose,CA,IEEE CS Press,USA,1988
[33]F.Berman and R.Wolski,The AppLes Project:A Status Report[C],Proceedings of the 8~(th) NEC Research Symposium,Berlin,Germany,May 1997
[34]M.Neary,A.Phipps,S.Richman,P.Cappello,Javelin 2.0:Java-Based Parallel Computing on the Internet[C],Proceedings of European Parallel Computing Conference(Euro-Par 2000),Germany,2000
[35]W.Hoschek,J.Jaen-Martinez,A.Samar,H.Stockinger,and K.Stockinger,Data Management in an International Data Grid Project[C],Proceedings of the First IEEE/ACM international Workshop on Grid Computing,(Springer Verlag Press,Germany),India,2000
[36]R.Buyya,D.Abramson,and J.Giddy,An Economy Driven Resource Management Architecture for Global Computational Power Grids[C],Proceedings of the 2000 International Conference on Parallel and Distributed Processing Techniques and Applications(PDPTA 2000),June 26-29,2000,Las Vegas,USA,CSREA Press,USA,2000
[37]D.Ferguson,C.Nikolaou,J.Sairamesh,and Y.Yemini,Economic Models for Allocating Resources in Computer Systems[C],In Market-based Control:a Paradigm for Distributed Resource Allocation,World Scientific Press,Singapore,1996
[38]D Abramson,I.Foster,J.Giddy,A.Lewis,R.Sosic,R.Sutherst,N.White,The Nimrod Computational Workbench:A Case Study in Desktop Metacomputing[J],Australian Computer Science Conference(ACSC 97),Macquarie University,Sydney,Feb 1997
[39]D.Abramson,R.Sosic,J.Giddy and B.Hall,Nimrod:A Tool for Performing Parameterized Simulations Using Distributed Workstations[C],Proceedings of the 4~(th) IEEE International Symposium on High Performance Distributed Computing,Virginia,August 1995,IEEE CS Press, USA,1995
[40]D.Abramson,P.Roe,L.Kotler,and D.Mather,ActiveSheets:Super-Computing with Spreadsheets[C].2001 High Performance Computing Symposium(HPC'01),Advanced Simulation Technologies Conference,April 2001
[41]G.Zacharia,P.Maes.Trust Management Through Reputation Mechanisms[J].Applied Artificial Intelligence Journal,2000,14(9):881-908
[42]B.Padovan,S.Sackmann,et al.A Prototype for an Agent-based Secure Electronic Marketplace including Reputation Tracking Mechanisms[J].International Journal of Electronic Commerce,2002,6(4):93
[43]马满福,吴健,陈丁剑,邓磊,胡正国。网格经济模型中基于信誉度的资源选择[J]。计算机工程,2006,32(17):175-177
[44]马满福,吴健,胡正国,陈丁剑,蒋芸。网格计算经济中的信誉度体系结构[J]。计算机应用,2005,25(4),940-943
[45]F.Azzedin,M.Maheswaran.Towards Trust-aware Resource Management in Grid Computing Systems[C].Cluster Computing and the Grid 2~(nd) IEEE/ACM International Symposium CC-Grid2002,May 21-24 2002,419-424