大容量VoIP系统关键技术研究与实现
|
摘要
VoIP技术的发展速度虽然很快,但建设和维护大容量VoIP系统依然是一项挑战性的任务。当前,如何构建大容量的VoIP系统已经成为了一个紧迫的课题。目前,系统扩容技术主要有两种:性能优化和集群。这两种扩容技术适用于不同等级的扩容需求。
论文在深入分析服务器系统性能优化关键技术和集群关键技术的基础上,分别从性能优化和集群这两方面入手,以西安普罗通信FreePP系统为课题来源,研究并实现了一种大容量VoIP系统。论文对所研究的VoIP系统的IAX2注册业务流访问数据库流程进行分析,结合排队论指出该系统的I/O瓶颈并且从性能优化角度提出与实现了一种基于数据库动态连接池的系统扩容方案。在所研究VoIP系统原架构的基础上,结合IAX2协议,设计与实现了一个采用分布式负载均衡方式的集群系统。在局域网环境下,论文对两种扩容方案进行仿真测试,结果初步证明:性能优化方案能将系统容量扩大至万级;集群方案能构建一个具有一定伸缩性和可用性的系统。
Although the development of VoIP technology is very rapid, the construction and maintenance of large-capacity VoIP systems remains a challenging task. Presently, how to build high-capacity VoIP system has become a pressing issue.Currently, there are two key technologies for system capacity enlargement: performance optimization and cluster. They apply to different levels of needs of system capacity enlargement.
On the basis of researching the key technologies of server system performance optimization and cluster, using the Xi'an Browan FreePP system as the subject source, this paper did researches and implemented a large capacity VoIP system from aspects of performance optimization and cluster separately. This paper analyzed the IAX2 registration service flow of the VoIP system researched, pointed out the I/O bottleneck of the system with queuing theory, proposed and implemented a performance optimization program based on the database dynamic connection pool .With the original architecture of the VoIP system researched and IAX2 protocol ,this paper designed and implemented a cluster system adopted a distributed load balancing mode. This paper made simulation testes on these two programs in the LAN environment. Through the result, it is proved that the performance optimization program can enlarge the system capacity to ten thousand level and the cluster program can build a system with some scalability and availability.
论文在深入分析服务器系统性能优化关键技术和集群关键技术的基础上,分别从性能优化和集群这两方面入手,以西安普罗通信FreePP系统为课题来源,研究并实现了一种大容量VoIP系统。论文对所研究的VoIP系统的IAX2注册业务流访问数据库流程进行分析,结合排队论指出该系统的I/O瓶颈并且从性能优化角度提出与实现了一种基于数据库动态连接池的系统扩容方案。在所研究VoIP系统原架构的基础上,结合IAX2协议,设计与实现了一个采用分布式负载均衡方式的集群系统。在局域网环境下,论文对两种扩容方案进行仿真测试,结果初步证明:性能优化方案能将系统容量扩大至万级;集群方案能构建一个具有一定伸缩性和可用性的系统。
Although the development of VoIP technology is very rapid, the construction and maintenance of large-capacity VoIP systems remains a challenging task. Presently, how to build high-capacity VoIP system has become a pressing issue.Currently, there are two key technologies for system capacity enlargement: performance optimization and cluster. They apply to different levels of needs of system capacity enlargement.
On the basis of researching the key technologies of server system performance optimization and cluster, using the Xi'an Browan FreePP system as the subject source, this paper did researches and implemented a large capacity VoIP system from aspects of performance optimization and cluster separately. This paper analyzed the IAX2 registration service flow of the VoIP system researched, pointed out the I/O bottleneck of the system with queuing theory, proposed and implemented a performance optimization program based on the database dynamic connection pool .With the original architecture of the VoIP system researched and IAX2 protocol ,this paper designed and implemented a cluster system adopted a distributed load balancing mode. This paper made simulation testes on these two programs in the LAN environment. Through the result, it is proved that the performance optimization program can enlarge the system capacity to ten thousand level and the cluster program can build a system with some scalability and availability.
引文
[1] PR.P.Swale . VoIP—Panacea or PIG's Ear . BT Technology Journal . Apr.2001,volume19(2) . 9-22
[2] Claudio Feijoo,Jose Luis Gomez-Barroso,David Rojo-Alonoso . A European perspective of VoIP in market competition . Communications of the ACM . Nov.2008,volume 51(11) . 118-120
[3] SrinivasanSeetharaman. IP over DWDM .http://www.cis.ohio-state.edu.Aug.2007
[4] Lucent Technologies .Web ProForum tutorial:DWDM . http://www.webproforum comlacrobatldwdm.pdf[4]. April 2006.126-145
[5] Microsoft Corporation.Resolving Network Bottlenecks .Microsoft TechNet. 2009
[6] Dennis de Champeaux,Doung Lea,Penelope Faure . Object-Oriented System Development . Addison Wesley,1993 . 102-105
[7] B?rje Lindh . Application Performance Optimization. Sun BluePrints OnLine . March 2002 . 1-5
[8]李彬,任国林.Linux内核基于对称多处理机的实现分析.计算机技术与发展.2006年,16(1) .130-131
[9]章文嵩.LVS中文文档.http://www.linuxvirtualserver.org/zh/lvs1.html. 2002.03
[10] Zhenghua Xue, Xiaoshe Dong, Junyang Li, etc. EISR:A Deployment System for Large-Scale Server Cluster.2008 Seventh International Conference on Grid and Cooperative Computing.Otc.2008. 563-568
[11] Zhenghua Xue, Xiaoshe Dong, Shengqun Fan. A Novel Management Architecture for Large-scale Server Cluster. CIT 2008.8th IEEE Internationsl Confeence. July 2008 . 273-278
[12] Zhang Lei,Dai Hao,Ma Mingka,etc. SCTP-Based Server Cluster Heartbeat Detection Mechanism. Computer Science and Computational Technology. December 2008, Volume 1 .118-122
[13] Schroeder, T.,Goddard, S.,Ramamurthy, B.Scalable Web server clustering technologies. Network,IEEE . May-June 2000,Volume14(3). 38-45
[14] Chun Zhang,Chang R.N.,Chang-shing Perng, etc. An Optimal Capacity Planning Algorithm for Provisioning Cluster-Based Failure-Resilient Composite Services. SCC '09 IEEE International Conference. September 2009. 112– 119
[15] Task Force on Cluster Computing. http://www.ieeetfcc.org/
[16] Mache Creeger.Cloud Computing: An Overview.Queue.June 2009,Volume 7(5).3-4
[17] Wikipedia .Cloud computing.http://en.wikipedia.org/wiki/Cloud_computing.2009
[18] Eric Knorr, Galen Gruman.What cloud computing really means. InfoWorld
[19] Sun Microsystems,Inc.Server Virtualization and Manageability.Websun.com. September 2004. 5-6
[20] Faouzi Kamoun. Virtualizing the Datacenter Without Compromising Server Performance.ACM Ubiquity.August 2009,Vol.2009(9).
[21] Asterisk.http://www.asterisk.org/
[22] M. Spencer, B. Capouch, E. Guy, etc. IAX: Inter-Asterisk eXchange Version 2. RFC 5456, IETF . 2009
[23]关宇平.基于C/S结构应用系统性能优化的方法的探讨.石油工业计算机应用.1999年1月.12-13.
[24]谭骏珊,吴昌盛.基于B/S模式应用系统性能优化的研究.计算机应用.2003年1月,voluem23(1) .70-72
[25]王敏.计算机性能优化的探讨.铁路计算机应用.1999年1月,Vol.12 No.9.170
[26] Rajat P. Garg, Ilya Sharapov. Techniques for Optimization Application:High Performance Compution.Prentice Hall Profession Technical Reference.2002.122-145
[27] William Stallings (美)著,陈渝译.操作系统精髓与设计原理.北京:电子工业出版社,2008年4月.122-124
[28] Jon Mountjoy,Avinash Chugh.Weblogic权威指南(影印版).南京:东南大学出版社,2005.312-356
[29]冯玉琳,黄涛,金蓓弘.网络分布计算和软件工程.北京:科学出版社,2003.230-235
[30] Rajkumar Buyya著,郑伟民,汪东升,石威等译.高性能集群计算:结构与系统(第一卷) .北京:电子工业出版社,2001年6月.130-132, 6-7
[31] K.Hamzeh,G.Pall,W.Verthein,etc.Point-to-Ponit Tunneling Protocol. RFC2637,IETF.1999
[32] Hui Chi-chung,Chanson S.T. Theoretical analysis of heterogeneous dynamic load-balancing problem using a hydrodynamic approach. Journal of Parallel and Distributed Computing,1997,43(2).139-146
[33] W.Richard Stevens(美),Stephen A.Rago(美)著,尤晋元、张亚英、戚正伟译.UNIX环境高级编程.第二版.北京:人民邮电出版社,2008年12月.122-132
[34]李建东、盛敏.通信网络基础.北京:高等教育出版社,2004年8月.p89-95
[2] Claudio Feijoo,Jose Luis Gomez-Barroso,David Rojo-Alonoso . A European perspective of VoIP in market competition . Communications of the ACM . Nov.2008,volume 51(11) . 118-120
[3] SrinivasanSeetharaman. IP over DWDM .http://www.cis.ohio-state.edu.Aug.2007
[4] Lucent Technologies .Web ProForum tutorial:DWDM . http://www.webproforum comlacrobatldwdm.pdf[4]. April 2006.126-145
[5] Microsoft Corporation.Resolving Network Bottlenecks .Microsoft TechNet. 2009
[6] Dennis de Champeaux,Doung Lea,Penelope Faure . Object-Oriented System Development . Addison Wesley,1993 . 102-105
[7] B?rje Lindh . Application Performance Optimization. Sun BluePrints OnLine . March 2002 . 1-5
[8]李彬,任国林.Linux内核基于对称多处理机的实现分析.计算机技术与发展.2006年,16(1) .130-131
[9]章文嵩.LVS中文文档.http://www.linuxvirtualserver.org/zh/lvs1.html. 2002.03
[10] Zhenghua Xue, Xiaoshe Dong, Junyang Li, etc. EISR:A Deployment System for Large-Scale Server Cluster.2008 Seventh International Conference on Grid and Cooperative Computing.Otc.2008. 563-568
[11] Zhenghua Xue, Xiaoshe Dong, Shengqun Fan. A Novel Management Architecture for Large-scale Server Cluster. CIT 2008.8th IEEE Internationsl Confeence. July 2008 . 273-278
[12] Zhang Lei,Dai Hao,Ma Mingka,etc. SCTP-Based Server Cluster Heartbeat Detection Mechanism. Computer Science and Computational Technology. December 2008, Volume 1 .118-122
[13] Schroeder, T.,Goddard, S.,Ramamurthy, B.Scalable Web server clustering technologies. Network,IEEE . May-June 2000,Volume14(3). 38-45
[14] Chun Zhang,Chang R.N.,Chang-shing Perng, etc. An Optimal Capacity Planning Algorithm for Provisioning Cluster-Based Failure-Resilient Composite Services. SCC '09 IEEE International Conference. September 2009. 112– 119
[15] Task Force on Cluster Computing. http://www.ieeetfcc.org/
[16] Mache Creeger.Cloud Computing: An Overview.Queue.June 2009,Volume 7(5).3-4
[17] Wikipedia .Cloud computing.http://en.wikipedia.org/wiki/Cloud_computing.2009
[18] Eric Knorr, Galen Gruman.What cloud computing really means. InfoWorld
[19] Sun Microsystems,Inc.Server Virtualization and Manageability.Websun.com. September 2004. 5-6
[20] Faouzi Kamoun. Virtualizing the Datacenter Without Compromising Server Performance.ACM Ubiquity.August 2009,Vol.2009(9).
[21] Asterisk.http://www.asterisk.org/
[22] M. Spencer, B. Capouch, E. Guy, etc. IAX: Inter-Asterisk eXchange Version 2. RFC 5456, IETF . 2009
[23]关宇平.基于C/S结构应用系统性能优化的方法的探讨.石油工业计算机应用.1999年1月.12-13.
[24]谭骏珊,吴昌盛.基于B/S模式应用系统性能优化的研究.计算机应用.2003年1月,voluem23(1) .70-72
[25]王敏.计算机性能优化的探讨.铁路计算机应用.1999年1月,Vol.12 No.9.170
[26] Rajat P. Garg, Ilya Sharapov. Techniques for Optimization Application:High Performance Compution.Prentice Hall Profession Technical Reference.2002.122-145
[27] William Stallings (美)著,陈渝译.操作系统精髓与设计原理.北京:电子工业出版社,2008年4月.122-124
[28] Jon Mountjoy,Avinash Chugh.Weblogic权威指南(影印版).南京:东南大学出版社,2005.312-356
[29]冯玉琳,黄涛,金蓓弘.网络分布计算和软件工程.北京:科学出版社,2003.230-235
[30] Rajkumar Buyya著,郑伟民,汪东升,石威等译.高性能集群计算:结构与系统(第一卷) .北京:电子工业出版社,2001年6月.130-132, 6-7
[31] K.Hamzeh,G.Pall,W.Verthein,etc.Point-to-Ponit Tunneling Protocol. RFC2637,IETF.1999
[32] Hui Chi-chung,Chanson S.T. Theoretical analysis of heterogeneous dynamic load-balancing problem using a hydrodynamic approach. Journal of Parallel and Distributed Computing,1997,43(2).139-146
[33] W.Richard Stevens(美),Stephen A.Rago(美)著,尤晋元、张亚英、戚正伟译.UNIX环境高级编程.第二版.北京:人民邮电出版社,2008年12月.122-132
[34]李建东、盛敏.通信网络基础.北京:高等教育出版社,2004年8月.p89-95