基于Whois的拓扑感知P2P研究
|
摘要
结构化对等网系统通常是建立在忽视底层网络的逻辑重叠网上的虚拟网络。基于分布式哈希表技术的对等网系统把关键字映射成ID标识符并以ID标识符构建重叠网。这种设计使结构化对等网在扩展性和容错性方面都有不错的性能,但也在通信效率和系统负载方面产生了诸多问题。
本文针对其中最突出的物理网和重叠网不匹配带来的高延时寻路、跨地区流量过多进行了重点研究,提出了一种地理感知重叠网(G-Chord),来构建含有位置信息的P2P网络。通过对Whois系统研究后,根据各ISP机构在互联网注册服务机构提交的网络注册信息中提取的关键信息并采用能统一显示地理位置及ISP归属信息的地理编码来构建IP地理信息数据库。本文提出了一种地理相关的标识符方法来构建重叠网,采用此方法后标识符前缀是通过哈希计算地理编码值来得到的,这样相同区域的节点拥有相同的标识符前缀从而在重叠网上也是相邻的。这种机制解决了对等网构建重叠网时出现的拓扑失配问题。
与现今已存在的拓扑匹配算法相比,本文提出的方法具有鲜明的几个优势:首先,实现拓扑感知的过程不产生额外通信量,减轻了网络负担。其次,算法设计合理,且不改变现有的P2P节点资源搜索机制,具有较高的可行性。通过实验模拟证明,该设计方法能明显降低跨域流量、降低通信延时。
Structured peer-to-peer systems are often built on the overlay network which neglects the underlying of the Internet. P2P systems which based on DHT technology map keys to ID identifier then construct a overlay network by using the identifier.This design makes structured peer-to-peer systems have good performance in expansibility and fault tolerance, however it also resultes in many problems on communication efficiency and system load.
In this paper , we focus on the problem of the high-delay route and trans-regional flow then we propose one topology-aware overlay network named G-Chord to structure a new geographic peer-to-peer network. Accoring to the information submited by ISP in internet registries, this thesis extract the key information and use a geocode which shows the geographic location and ownership information of ISP uniformly to build IP geographic information database.The thesis raises a geographic identifier method to build overlay network. In this method, we get prefix identifier value by hash the geocode value and it makes nodes in the same district get the same prefix identifier then they are neighbours in the overlay network. It solves the problem of topology mismatch.
Comparing with current algorithms, the algorithm in the etheis has several significant advantages: firstly, the process of achieve topology-aware does not generate additional traffic and reduces burden of network. Secondly , algorithm is reasonable and does not change the P2P resource search mechanism.It has high feasibility. Simulation experiments show that the design method can significantly reduce the cross-border traffic, reduce the communication delay.
本文针对其中最突出的物理网和重叠网不匹配带来的高延时寻路、跨地区流量过多进行了重点研究,提出了一种地理感知重叠网(G-Chord),来构建含有位置信息的P2P网络。通过对Whois系统研究后,根据各ISP机构在互联网注册服务机构提交的网络注册信息中提取的关键信息并采用能统一显示地理位置及ISP归属信息的地理编码来构建IP地理信息数据库。本文提出了一种地理相关的标识符方法来构建重叠网,采用此方法后标识符前缀是通过哈希计算地理编码值来得到的,这样相同区域的节点拥有相同的标识符前缀从而在重叠网上也是相邻的。这种机制解决了对等网构建重叠网时出现的拓扑失配问题。
与现今已存在的拓扑匹配算法相比,本文提出的方法具有鲜明的几个优势:首先,实现拓扑感知的过程不产生额外通信量,减轻了网络负担。其次,算法设计合理,且不改变现有的P2P节点资源搜索机制,具有较高的可行性。通过实验模拟证明,该设计方法能明显降低跨域流量、降低通信延时。
Structured peer-to-peer systems are often built on the overlay network which neglects the underlying of the Internet. P2P systems which based on DHT technology map keys to ID identifier then construct a overlay network by using the identifier.This design makes structured peer-to-peer systems have good performance in expansibility and fault tolerance, however it also resultes in many problems on communication efficiency and system load.
In this paper , we focus on the problem of the high-delay route and trans-regional flow then we propose one topology-aware overlay network named G-Chord to structure a new geographic peer-to-peer network. Accoring to the information submited by ISP in internet registries, this thesis extract the key information and use a geocode which shows the geographic location and ownership information of ISP uniformly to build IP geographic information database.The thesis raises a geographic identifier method to build overlay network. In this method, we get prefix identifier value by hash the geocode value and it makes nodes in the same district get the same prefix identifier then they are neighbours in the overlay network. It solves the problem of topology mismatch.
Comparing with current algorithms, the algorithm in the etheis has several significant advantages: firstly, the process of achieve topology-aware does not generate additional traffic and reduces burden of network. Secondly , algorithm is reasonable and does not change the P2P resource search mechanism.It has high feasibility. Simulation experiments show that the design method can significantly reduce the cross-border traffic, reduce the communication delay.
引文
[1]王谐.P2P技术对电信行业的挑战及监管对策综述.电信快报,2006,(01):20-25
[2]Yarali,A Cherry.Internet Protocol Television(IPTV).2005
[3]Hu A.Video-on-demand broadcasting protocols:a comprehensive study.INFOCOM 2001,6(1):508-517
[4]汪莉娟,武文忠.基于智能节点的P2P重叠网NMS的设计与实现:[硕士学位论文].兰州:兰州大学,2005
[5]唐明董,张国清.P2P流量优化技术综述电信网技术,2009,9(01):1-7
[6]侯自强.NGN业务模型变化和电信变革.电信科学,2005,21(6):1-4
[7]魏亮.从设计理念看下一代因特网.电信技术,2006,(07):15-17
[8]陈建华,黄道颖等.计算机对等网络.P2P技术计算机工程与应用,2003,12(33):72-73
[9]Napster website,http://www.napster.com.1999
[10]S M Hotr,Routing information organization to support scalable interdomain routing with heterogeneous path requirements.University of Southern California,1994
[11]Ng T S E,Hui Zhang.Predicting Internet network distance with coordinates based approaches.INFOCOM,2002:112-121
[12]赵恒,陈杰.P2P技术的应用及其研究现状.电信快报,2004,9(3):42-48
[13]杨明川.P2P运营模式及P2P业务发展策略.中兴通讯技术,2007,9(06):5-7
[14]Y Chawathe,S Ramabhadran.A case study in building layered DHT applications.ACM SIGCOMM,2005.97-108
[15]任戬,阮秋琦.网络拓扑发现研究及网管模块实现:[硕士学位论文].北京:北京交通大学,2006
[16]William Stallings.SNMP网络管理.北京:中国电力出版,2001.140-152
[17]Miguel Castro,PeterDruchel Y.Exploiting network proximity in peer-to-peer overlay networks.Technical Report MSR-TR-2002-82,Microsoft Research,2002.211-232
[18]张联峰,刘乃安.综述:对等网(P2P)技术.计算机工程与应用,2003,39(12):142-148
[19]丛亮.我国宽带发展的挑战与机遇浅析.2007,20(7):50-54
[20]杨山河.实例解说Whois.2002,10(5):39-44
[21]Marcel Wald Vogel.Efficient topology-aware overlay network.ACMSIGCOM,2003,33(1):101-106
[22]IonStoiea,RobertMorris,DavidKarger,et al.Chord:A Scalable Peer-to-Peer Lookup Service for Internet Applications.MIT Laboratory for Computer Science.SIGCOMM,2001.78-91
[23]温忠智,高阶.Chord:一种新型P2P查找策略:[硕士学位论文].成都:四川大学,2005
[24]F Dabek,E Brunskill,MF Kaashoek,et al.Building peer-to-peer systems with Chord,a distributed lookup service.Proceedings of the 8th Workshop on Hot Topics in Operating System,2001.83-101
[25]Sylvia Ratnasamy,Paul Francis,Mark Handley,et al.A scalable content-addressable network.ACMSIGCOMM Conference,San Diego,31(2),2001:161-172
[26]BY Zhao,J Kubiatowicz,AD Joseph.Tapestry:An infrastructure for fault-tolerant wide-area location and routing.Proceedings of the ACMSIGCOMM,2001.75-91
[27]A Rowstron,P Druschel.Pastry:Scalable,Distributed object location and routing for large-scale peer-to-peer systems.IFIP/ACM International Conference on Distributed Systems Platforms,2001.45-67
[28]Pete Keleher,Samrat Bhattacharjee,Bujor Silaghi.Are virtualized overlay networks too much of a good thing? In Proceedings of International Workshop on Peer-to-Peer Systems,2002.78-90
[29]T S E Ng,H Zhang.Predicting Internet network distance with coordinates-based approaches.In Proceedings of the IEEE INFOCOM,2002.141-156
[30]M Waldvogel,R Rinaldi.Effieient topology-aware overlay network.ACM SIGC-OMM Computer Communication Review,2003.101-107
[31]M Pias,J Crowcroft,S Wilbur,T Harris,et al.Lighthouses for Scalable distri- buted location.Lecture notes in computer science,2003
[32]F Dabek,R Cox,F Kaashoek,et al.R Morris.Vivaldi:A decentralized network coordinate system.ACM SIGCOMM Computer Communication Review,2004.100-112
[33]L Tang,M Crovella.Virtual landmarks for the internet.Proceedings of the 3rd ACM SIGCOMM conference on Internet measurement,2003.143-152
[34]H Lim,J Hou,C-H Choi.Constructing internet coordinate system based On delay measurement.Internet Measurement Conference,2003.129-142
[35]Shansi Ren,Lei Guo,Song Jiang,et al.SAT-Match:A Self-Adaptive Topology Matching Method to Achieve Low Lookup Latency in Structured P2P Overlay Networks.Parallel and Distributed Processing Symposium Proceedings 18th International.2004.83-91
[36]黄继海,黄建华,毛军鹏等.拓扑探测技术在P2P媒体分发系统中的应用.电视技术,31(10):89-94
[37]Rekhter Y,T Li.A Border Gateway Protocol 4(BGP-4).Work in Progress Watson Research Center IBM Corp cisco Systems,1992
[38]L Garees-Eriee,K W Ross,EW Biersaek.Topology-Centric look-up Sevrice.in Proceedings of COST264/ACM Fifth Intenrational Workshop on Networked Group Communications,2003.79-88
[39]Merit Network.Internet Peroformance Measurement and Analysis(IPMA)Project.www.mcrit.edu/ipma/routingtable,2004
[40]中国教育与科研计算机网.www.edu.cn.2008
[41]SallyFloyd,VemPaxson.DifficultiesinsimulatingtheIntemet.IEEE/ACM TransactionsOn Networking,2001,9(4):392-403
[42]Ken Harrenstien,Vic White.RFC812- NICNAME/Whois.Network Information Center SRI International,1982
[43]刘辑.中国通信网络规模容量已跃居世界第一位.第四届中国信息港论坛,2002
[44]中国互联网信息中心.中国互联网络发展状况统计报告.www.cnnic.net,2008
[45]中华人民共和国国家统计局.最新县及县以上行政区划代码,2007
[46]Per-Ake Larson,Univ of Waterloo,Waterloo.Dynamic hash tables.Communications of the ACM,1988,31(4):446-457
[47]R Rivest.RFC1321:The MD5 Message-Digest Algorithm.Portal.acm.org,United States,1992
[48]D Eastlake,P Jones.RFC 3174:US secure hashing algorithm 1(SHA1).In Proceedings of the IEEE INFOCOM,2001
[49]李宪民.中国互联网的发展状况及趋势.情报杂志,2001,20(8):26-40
[50]刘瑜.内容分发网中基于对等网的流媒体分发技术研究:[博士学位论文].武汉:华中科技大学,2008
[51]Brite.A network topology generator,http://cs.bu.edu/brite/
[52]The Network Simulator(ns).www.isi.edu/nsnam/ns,2003
[53]Scalable Simulation Framework(SSF).www.ssfnet.org,2004
[54]B Waxman.Routing of Multipoint Connections.Selected Areas in Communications,IEEEE Journal,1988,6(9):1617-1622
[2]Yarali,A Cherry.Internet Protocol Television(IPTV).2005
[3]Hu A.Video-on-demand broadcasting protocols:a comprehensive study.INFOCOM 2001,6(1):508-517
[4]汪莉娟,武文忠.基于智能节点的P2P重叠网NMS的设计与实现:[硕士学位论文].兰州:兰州大学,2005
[5]唐明董,张国清.P2P流量优化技术综述电信网技术,2009,9(01):1-7
[6]侯自强.NGN业务模型变化和电信变革.电信科学,2005,21(6):1-4
[7]魏亮.从设计理念看下一代因特网.电信技术,2006,(07):15-17
[8]陈建华,黄道颖等.计算机对等网络.P2P技术计算机工程与应用,2003,12(33):72-73
[9]Napster website,http://www.napster.com.1999
[10]S M Hotr,Routing information organization to support scalable interdomain routing with heterogeneous path requirements.University of Southern California,1994
[11]Ng T S E,Hui Zhang.Predicting Internet network distance with coordinates based approaches.INFOCOM,2002:112-121
[12]赵恒,陈杰.P2P技术的应用及其研究现状.电信快报,2004,9(3):42-48
[13]杨明川.P2P运营模式及P2P业务发展策略.中兴通讯技术,2007,9(06):5-7
[14]Y Chawathe,S Ramabhadran.A case study in building layered DHT applications.ACM SIGCOMM,2005.97-108
[15]任戬,阮秋琦.网络拓扑发现研究及网管模块实现:[硕士学位论文].北京:北京交通大学,2006
[16]William Stallings.SNMP网络管理.北京:中国电力出版,2001.140-152
[17]Miguel Castro,PeterDruchel Y.Exploiting network proximity in peer-to-peer overlay networks.Technical Report MSR-TR-2002-82,Microsoft Research,2002.211-232
[18]张联峰,刘乃安.综述:对等网(P2P)技术.计算机工程与应用,2003,39(12):142-148
[19]丛亮.我国宽带发展的挑战与机遇浅析.2007,20(7):50-54
[20]杨山河.实例解说Whois.2002,10(5):39-44
[21]Marcel Wald Vogel.Efficient topology-aware overlay network.ACMSIGCOM,2003,33(1):101-106
[22]IonStoiea,RobertMorris,DavidKarger,et al.Chord:A Scalable Peer-to-Peer Lookup Service for Internet Applications.MIT Laboratory for Computer Science.SIGCOMM,2001.78-91
[23]温忠智,高阶.Chord:一种新型P2P查找策略:[硕士学位论文].成都:四川大学,2005
[24]F Dabek,E Brunskill,MF Kaashoek,et al.Building peer-to-peer systems with Chord,a distributed lookup service.Proceedings of the 8th Workshop on Hot Topics in Operating System,2001.83-101
[25]Sylvia Ratnasamy,Paul Francis,Mark Handley,et al.A scalable content-addressable network.ACMSIGCOMM Conference,San Diego,31(2),2001:161-172
[26]BY Zhao,J Kubiatowicz,AD Joseph.Tapestry:An infrastructure for fault-tolerant wide-area location and routing.Proceedings of the ACMSIGCOMM,2001.75-91
[27]A Rowstron,P Druschel.Pastry:Scalable,Distributed object location and routing for large-scale peer-to-peer systems.IFIP/ACM International Conference on Distributed Systems Platforms,2001.45-67
[28]Pete Keleher,Samrat Bhattacharjee,Bujor Silaghi.Are virtualized overlay networks too much of a good thing? In Proceedings of International Workshop on Peer-to-Peer Systems,2002.78-90
[29]T S E Ng,H Zhang.Predicting Internet network distance with coordinates-based approaches.In Proceedings of the IEEE INFOCOM,2002.141-156
[30]M Waldvogel,R Rinaldi.Effieient topology-aware overlay network.ACM SIGC-OMM Computer Communication Review,2003.101-107
[31]M Pias,J Crowcroft,S Wilbur,T Harris,et al.Lighthouses for Scalable distri- buted location.Lecture notes in computer science,2003
[32]F Dabek,R Cox,F Kaashoek,et al.R Morris.Vivaldi:A decentralized network coordinate system.ACM SIGCOMM Computer Communication Review,2004.100-112
[33]L Tang,M Crovella.Virtual landmarks for the internet.Proceedings of the 3rd ACM SIGCOMM conference on Internet measurement,2003.143-152
[34]H Lim,J Hou,C-H Choi.Constructing internet coordinate system based On delay measurement.Internet Measurement Conference,2003.129-142
[35]Shansi Ren,Lei Guo,Song Jiang,et al.SAT-Match:A Self-Adaptive Topology Matching Method to Achieve Low Lookup Latency in Structured P2P Overlay Networks.Parallel and Distributed Processing Symposium Proceedings 18th International.2004.83-91
[36]黄继海,黄建华,毛军鹏等.拓扑探测技术在P2P媒体分发系统中的应用.电视技术,31(10):89-94
[37]Rekhter Y,T Li.A Border Gateway Protocol 4(BGP-4).Work in Progress Watson Research Center IBM Corp cisco Systems,1992
[38]L Garees-Eriee,K W Ross,EW Biersaek.Topology-Centric look-up Sevrice.in Proceedings of COST264/ACM Fifth Intenrational Workshop on Networked Group Communications,2003.79-88
[39]Merit Network.Internet Peroformance Measurement and Analysis(IPMA)Project.www.mcrit.edu/ipma/routingtable,2004
[40]中国教育与科研计算机网.www.edu.cn.2008
[41]SallyFloyd,VemPaxson.DifficultiesinsimulatingtheIntemet.IEEE/ACM TransactionsOn Networking,2001,9(4):392-403
[42]Ken Harrenstien,Vic White.RFC812- NICNAME/Whois.Network Information Center SRI International,1982
[43]刘辑.中国通信网络规模容量已跃居世界第一位.第四届中国信息港论坛,2002
[44]中国互联网信息中心.中国互联网络发展状况统计报告.www.cnnic.net,2008
[45]中华人民共和国国家统计局.最新县及县以上行政区划代码,2007
[46]Per-Ake Larson,Univ of Waterloo,Waterloo.Dynamic hash tables.Communications of the ACM,1988,31(4):446-457
[47]R Rivest.RFC1321:The MD5 Message-Digest Algorithm.Portal.acm.org,United States,1992
[48]D Eastlake,P Jones.RFC 3174:US secure hashing algorithm 1(SHA1).In Proceedings of the IEEE INFOCOM,2001
[49]李宪民.中国互联网的发展状况及趋势.情报杂志,2001,20(8):26-40
[50]刘瑜.内容分发网中基于对等网的流媒体分发技术研究:[博士学位论文].武汉:华中科技大学,2008
[51]Brite.A network topology generator,http://cs.bu.edu/brite/
[52]The Network Simulator(ns).www.isi.edu/nsnam/ns,2003
[53]Scalable Simulation Framework(SSF).www.ssfnet.org,2004
[54]B Waxman.Routing of Multipoint Connections.Selected Areas in Communications,IEEEE Journal,1988,6(9):1617-1622