P2P应用对网络的影响和识别方法研究
|
摘要
P2P技术在出现后的短短几年间得到了迅速发展,P2P应用已经成为当今互联网的主流,根据目前的统计数据显示,P2P流量已经占据了互联网总流量的60%—70%。P2P应用的迅猛发展对网络带来了巨大的负面影响,占用了过多的网络带宽,恶化了网络的安全环境。为了管理P2P应用,首先需要识别P2P流量,因此,准确地识别P2P流量对于保证网络的服务质量、流量监控、网络安全有着重大意义。
本文对目前存在的P2P流量识别方法进行了分类归纳,从正确性、精确性、算法的复杂程度三方面对它们的识别性能进行了对比,发现基于端口的流量识别方法能够识别的P2P应用数量有限,且准确率在不断下降;基于P2P流量特征的识别方法现在还不能对P2P流量进行分类识别,还需要进一步完善;基于特征码的识别方法,不但能够实现对P2P流量的准确识别和应用分类,而且实现技术简单,广泛地应用于各种流量检测系统。
对于基于特征码的P2P流量识别技术,对P2P的分析是其前提和基础。本文分别分析了P2P的定义、网络拓扑结构、节点资源搜索方式以及主要应用,总结了P2P流量的特点和对网络的影响。通过对迅雷、BitTorrent、PPlive这三种应用最广泛的P2P应用软件的工作过程进行分析,归纳出它们的特征码,并以此总结出一些获取特征码的方法。
本文最后利用基于特征码的P2P流控软件Panabit,对一个发生网络拥塞的校园网进行部署,发现P2P应用是造成网络拥塞的主要原因。通过制定相应的优化策略,取得了满意的效果,同时也为相关的案例提供了一定的借鉴作用。
After the emergence of P2P technology in the short few years has been rapid development,P2P applicatins have become the mainstream of today's internet,based on current statics,P2P traffic have occupied 60%-70% of the total traffic.The rapid growth of P2P application on the network has brought enormous negative impact,take up too much network bandwidth,deteriorated network'security.In order to manage P2P applicatons,first of all need to identify P2P traffic,therefore,accurate identification of P2P traffic to guarantee network service quality,traffic monitoring,network security is of great significance.
The existing P2P traffic identification methods are classified in this paper, From the correctness, accuracy, algorithm complexity, compared to their recognition performance. Found that the flow of port-based identification method can identify a limited number of P2P applications, And the steady decline in accuracy; Feature-based P2P traffic identification method is not yet on the P2P traffic classification, Require further improvement; Signature-based identification method, not only to achieve an accurate identification of P2P traffic and application classification, but also implementation technology is simple, widely used in a variety of flow measurement and control system
For signature-based P2P traffic identification technology, P2P protocol must be analyzed. This paper analyzes the P2P technology works, network topology and node resource search. Summarizes the characteristics of P2P traffic. Through BitTorrent, Thunder, PPlive the three most widely used P2P applications work process analysis, Sum up their signature, and sums up some method of obtaining signatures.
Finally, use flow control software Panabit, to deploy a network congestion occurs in the Campus Network, found that network congestion caused by P2P applications is the main reason. Through the development of appropriate optimization strategies, and achieved satisfactory results, As well as give relevant case provides some reference.
本文对目前存在的P2P流量识别方法进行了分类归纳,从正确性、精确性、算法的复杂程度三方面对它们的识别性能进行了对比,发现基于端口的流量识别方法能够识别的P2P应用数量有限,且准确率在不断下降;基于P2P流量特征的识别方法现在还不能对P2P流量进行分类识别,还需要进一步完善;基于特征码的识别方法,不但能够实现对P2P流量的准确识别和应用分类,而且实现技术简单,广泛地应用于各种流量检测系统。
对于基于特征码的P2P流量识别技术,对P2P的分析是其前提和基础。本文分别分析了P2P的定义、网络拓扑结构、节点资源搜索方式以及主要应用,总结了P2P流量的特点和对网络的影响。通过对迅雷、BitTorrent、PPlive这三种应用最广泛的P2P应用软件的工作过程进行分析,归纳出它们的特征码,并以此总结出一些获取特征码的方法。
本文最后利用基于特征码的P2P流控软件Panabit,对一个发生网络拥塞的校园网进行部署,发现P2P应用是造成网络拥塞的主要原因。通过制定相应的优化策略,取得了满意的效果,同时也为相关的案例提供了一定的借鉴作用。
After the emergence of P2P technology in the short few years has been rapid development,P2P applicatins have become the mainstream of today's internet,based on current statics,P2P traffic have occupied 60%-70% of the total traffic.The rapid growth of P2P application on the network has brought enormous negative impact,take up too much network bandwidth,deteriorated network'security.In order to manage P2P applicatons,first of all need to identify P2P traffic,therefore,accurate identification of P2P traffic to guarantee network service quality,traffic monitoring,network security is of great significance.
The existing P2P traffic identification methods are classified in this paper, From the correctness, accuracy, algorithm complexity, compared to their recognition performance. Found that the flow of port-based identification method can identify a limited number of P2P applications, And the steady decline in accuracy; Feature-based P2P traffic identification method is not yet on the P2P traffic classification, Require further improvement; Signature-based identification method, not only to achieve an accurate identification of P2P traffic and application classification, but also implementation technology is simple, widely used in a variety of flow measurement and control system
For signature-based P2P traffic identification technology, P2P protocol must be analyzed. This paper analyzes the P2P technology works, network topology and node resource search. Summarizes the characteristics of P2P traffic. Through BitTorrent, Thunder, PPlive the three most widely used P2P applications work process analysis, Sum up their signature, and sums up some method of obtaining signatures.
Finally, use flow control software Panabit, to deploy a network congestion occurs in the Campus Network, found that network congestion caused by P2P applications is the main reason. Through the development of appropriate optimization strategies, and achieved satisfactory results, As well as give relevant case provides some reference.
引文
[1]贺思德,申浩如.计算机网络安全与应用.科学出版社[M]. 2007.
[2] Cachelogic中国互联网流量分析报告[EB/OL]. http://www.cachelogic.com/home/pages/research/p2p2005.php.
[3] Subhabrata S,Spatscheck O,Wang D. Accurate scalable in-network identification of p2p traffic using application signatures[J]. Proceeding of the 13 th Internet Conference on Word Wide Web. 2004:512-521
[4] Panabit官方网站. http://www.panabit.com/。
[5]华为sig .http://www.huawei.com.
[6]城市热点.http://www.doctorcom.com.
[7]思科中国.http://www.cisco.com.cn.
[8] Bandwidth Management Solutions.http://www.allot.com.
[9] M.Harren,J.M.Hellerstein,R.huebschand.Complex Queries in DHT based Peer-to-Peer Networks[J].The first International Workshop on Peer-to-Peer Systems.2002,(3):242-259.
[10] I.Stoics,R.Morris,D.Karger.Chord:A Scalable Peer-to-Peer Lookup Service for Internet Applications[J].ACM SIGCOMM.2001,(01):161-172.
[11] A.RowStron,P.Druschel.Pastry:Scalable,Distributed Object Loacation and routing for Large-scale Peer-to-Peer Systems[J].IFIP/ACM International Conference on Distributed Systems Platforms.2001,(1):329-350.
[12] Tylvia Ratnasamy,Paul Francis,Mark Handley.A Scable Content-addressable Network[J].ACM SIGCOMM.2001,(1):161-172.
[13] B.Zhao,J.D.Kubiatowicz,A.D.Joseph.Tapestry:An Infrastructure for Fault-tolerant Wide-area Location and Routing[J].Technical report,UCB/CSD-01,1141,UC Berkeley.April 2001.
[14] Yang B,Garcia M H.Improving Search in Peer-to-Peer Networks[J].the 22nd IEEE International Conference on Distrituted Computing.2002.
[15] Kalogeraki V,Gunopulos D,Yazti Z D.A Local Search Mechanism for P2P Networks[J].CIKM’02.2002,(11):4-9.
[16] Gkantsidis C,Mihail M,Saberi A Random Walks in Peer-to-Peer Networks[J].IEEE Infocom.2004,(4).
[17] Subhabrata S,Wang J. A nalyzing peer-to-peer traffic across large network[J]. The 2 nd Ineternet measurement workshop. 2004: 219-232.
[18]倪冰,陈云清. P2P流量监控技术的研究及现网试验[J].吉林大学学报(信息科学版). 2009.
[19] T.Karagiannis, A.Broido, N.Brownlee, Kc Claffy, and M.Faloutsos. Is p2p dying or justhiding?[J]. In Globecom,Dallas,TX,USA,November. 2004.
[20] HUI Liu,WEN Feng,et al.A peer-to-peer traffic identification method using machine learning[J]. Intrnational Conference on Networking,Architecture and Storage.2007:155-160.
[21] Thomas K, Broido A, Faloutsos M. Trasport layer identification of p2p traffic[J]. Proceedings of the 4 th ACM SIGCOMM conferrence on intenet measurment italy. 2004:121-134.
[22] Constantinou F,Mavrommatis P. Identifying known and unknown Peer-to-Peer traffic[J].Fifth IEEE Internetional symposium on network computing and application cambridge. 2006:93-102.
[23] Wireshark.http://www.wireshark.org.
[24]罗杰文.P2P:互联网时代的计算革命[N].计算机世界报,2006-09-20.
[25]刘文涛.网络安全编程技术与实例[M].机械工业出版社.2008.
[26]余浩,徐明伟. P2P流检技术研究综述[J].清华大学学报(自然科学版).2009年第49卷第4期.
[27]刘斌,李之唐,李佳.一种基于流特征的P2P流量实时识别方法[J].厦门大学学报(自然科学版).2007.
[28]徐鹤,王汝传.一种P2P流量监控系统的设计及实现[J].计算机技术与发展. 2009.
[29]吴敏,王汝传.基于主机的P2P流量检测与控制方案[J].计算机技术与发展. 2009.
[30]孙美凤,张赟,陈云箐,陈经涛.基于流量模式的P2P流量识别方法综述[J].计算机应用研究. 2009.
[31]陈忠平.五联网P2P流量的影响与控制[J].铁道通信信号. 2009.
[32]陈姝英.迅雷P2SP架构及服务策略的分析[D].北京交通大学.硕士论文. 2007.
[33]赵瑞.基于特征串的P2P流量识别研究与实现[D].电子科技大学.硕士论文. 2009.
[34]王正旭. P2P流量可控特征提取与识别技术[D].天津理工大学.硕士论文. 2009.
[35]张文,赵子铭. P2P网络技术原理与C++开发案例[M].人民邮电出版社. 2008.
[36]邢小良. P2P技术及其应用[M].人民邮电出版社. 2008.
[37]杨岳湘,李强.基于双重特征的P2P流量检测方法[J].通信学报. 2006:134~138.
[38] Plissonneau L, Costeux J L, Brown P. Analysis of Peer-to-Peer Traffic on ADSL[J]. In PAM2005,volume 3431 of LNCS Springer. 2005:69～82.
[39] Moore A W, Papagiannaki K. Toward the Accurate Identification of Network Applications[C].In P AM2005.Boston,MA,2005,41～54.
[40] Pouwelseja J, Garbacki P. The bittorrent P2P file-sharing system:measurements and analysis[J]. Proceedings of the international workshop on peer-to-peer systems, 2005, 123-129.
[41]王孝伟. P2P流量的识别与检测技术的研究[D].重庆大学.硕士论文.2008.
[42]张文,侯立东.基于传输层标志位的P2P流量识别技术[J].科技咨询.2007,(01):163-164.
[43]徐恪,吴建平,徐明伟.高等计算机网络—体系结构、协议机制、算法设计与路由器技术[M].机械工业出版社.2009.
[2] Cachelogic中国互联网流量分析报告[EB/OL]. http://www.cachelogic.com/home/pages/research/p2p2005.php.
[3] Subhabrata S,Spatscheck O,Wang D. Accurate scalable in-network identification of p2p traffic using application signatures[J]. Proceeding of the 13 th Internet Conference on Word Wide Web. 2004:512-521
[4] Panabit官方网站. http://www.panabit.com/。
[5]华为sig .http://www.huawei.com.
[6]城市热点.http://www.doctorcom.com.
[7]思科中国.http://www.cisco.com.cn.
[8] Bandwidth Management Solutions.http://www.allot.com.
[9] M.Harren,J.M.Hellerstein,R.huebschand.Complex Queries in DHT based Peer-to-Peer Networks[J].The first International Workshop on Peer-to-Peer Systems.2002,(3):242-259.
[10] I.Stoics,R.Morris,D.Karger.Chord:A Scalable Peer-to-Peer Lookup Service for Internet Applications[J].ACM SIGCOMM.2001,(01):161-172.
[11] A.RowStron,P.Druschel.Pastry:Scalable,Distributed Object Loacation and routing for Large-scale Peer-to-Peer Systems[J].IFIP/ACM International Conference on Distributed Systems Platforms.2001,(1):329-350.
[12] Tylvia Ratnasamy,Paul Francis,Mark Handley.A Scable Content-addressable Network[J].ACM SIGCOMM.2001,(1):161-172.
[13] B.Zhao,J.D.Kubiatowicz,A.D.Joseph.Tapestry:An Infrastructure for Fault-tolerant Wide-area Location and Routing[J].Technical report,UCB/CSD-01,1141,UC Berkeley.April 2001.
[14] Yang B,Garcia M H.Improving Search in Peer-to-Peer Networks[J].the 22nd IEEE International Conference on Distrituted Computing.2002.
[15] Kalogeraki V,Gunopulos D,Yazti Z D.A Local Search Mechanism for P2P Networks[J].CIKM’02.2002,(11):4-9.
[16] Gkantsidis C,Mihail M,Saberi A Random Walks in Peer-to-Peer Networks[J].IEEE Infocom.2004,(4).
[17] Subhabrata S,Wang J. A nalyzing peer-to-peer traffic across large network[J]. The 2 nd Ineternet measurement workshop. 2004: 219-232.
[18]倪冰,陈云清. P2P流量监控技术的研究及现网试验[J].吉林大学学报(信息科学版). 2009.
[19] T.Karagiannis, A.Broido, N.Brownlee, Kc Claffy, and M.Faloutsos. Is p2p dying or justhiding?[J]. In Globecom,Dallas,TX,USA,November. 2004.
[20] HUI Liu,WEN Feng,et al.A peer-to-peer traffic identification method using machine learning[J]. Intrnational Conference on Networking,Architecture and Storage.2007:155-160.
[21] Thomas K, Broido A, Faloutsos M. Trasport layer identification of p2p traffic[J]. Proceedings of the 4 th ACM SIGCOMM conferrence on intenet measurment italy. 2004:121-134.
[22] Constantinou F,Mavrommatis P. Identifying known and unknown Peer-to-Peer traffic[J].Fifth IEEE Internetional symposium on network computing and application cambridge. 2006:93-102.
[23] Wireshark.http://www.wireshark.org.
[24]罗杰文.P2P:互联网时代的计算革命[N].计算机世界报,2006-09-20.
[25]刘文涛.网络安全编程技术与实例[M].机械工业出版社.2008.
[26]余浩,徐明伟. P2P流检技术研究综述[J].清华大学学报(自然科学版).2009年第49卷第4期.
[27]刘斌,李之唐,李佳.一种基于流特征的P2P流量实时识别方法[J].厦门大学学报(自然科学版).2007.
[28]徐鹤,王汝传.一种P2P流量监控系统的设计及实现[J].计算机技术与发展. 2009.
[29]吴敏,王汝传.基于主机的P2P流量检测与控制方案[J].计算机技术与发展. 2009.
[30]孙美凤,张赟,陈云箐,陈经涛.基于流量模式的P2P流量识别方法综述[J].计算机应用研究. 2009.
[31]陈忠平.五联网P2P流量的影响与控制[J].铁道通信信号. 2009.
[32]陈姝英.迅雷P2SP架构及服务策略的分析[D].北京交通大学.硕士论文. 2007.
[33]赵瑞.基于特征串的P2P流量识别研究与实现[D].电子科技大学.硕士论文. 2009.
[34]王正旭. P2P流量可控特征提取与识别技术[D].天津理工大学.硕士论文. 2009.
[35]张文,赵子铭. P2P网络技术原理与C++开发案例[M].人民邮电出版社. 2008.
[36]邢小良. P2P技术及其应用[M].人民邮电出版社. 2008.
[37]杨岳湘,李强.基于双重特征的P2P流量检测方法[J].通信学报. 2006:134~138.
[38] Plissonneau L, Costeux J L, Brown P. Analysis of Peer-to-Peer Traffic on ADSL[J]. In PAM2005,volume 3431 of LNCS Springer. 2005:69～82.
[39] Moore A W, Papagiannaki K. Toward the Accurate Identification of Network Applications[C].In P AM2005.Boston,MA,2005,41～54.
[40] Pouwelseja J, Garbacki P. The bittorrent P2P file-sharing system:measurements and analysis[J]. Proceedings of the international workshop on peer-to-peer systems, 2005, 123-129.
[41]王孝伟. P2P流量的识别与检测技术的研究[D].重庆大学.硕士论文.2008.
[42]张文,侯立东.基于传输层标志位的P2P流量识别技术[J].科技咨询.2007,(01):163-164.
[43]徐恪,吴建平,徐明伟.高等计算机网络—体系结构、协议机制、算法设计与路由器技术[M].机械工业出版社.2009.