P2P-VOD缓存策略的研究与仿真
|
摘要
数据缓存策略是流媒体软件播放质量的关键技术之一,当前流媒体播放软件的缓存策略采用的是SVC(Single video cache),该策略缓存的数据是当前节点正在观看的视频数据,然而当前的视频点播系统中,用户的观看行为成Zipf分布,系统中大部分的用户点播的都是当前最热门的视频,而系统中还存在大量的冷门视频,由于SVC缓存的是当前用户正在观看的视频,那么热门视频的数据副本量网络中缓存相对比较多,而冷门节点的数据副本量在系统中则缓存较少,观看冷门视频的节点,由于在网络中的缓存量少,就会存在视频播放不流畅,对服务器频繁请求数据的情况,正是因为存在这样的问题,所以本文针对此提出了一种P2P-VOD系统中的多视频缓存策略(MVC),该策略在分析现有SVC缓存策略的基础上,针对P2P视频点播系统的特点,为了改进流媒体系统视频播放的流畅程度,服务器负载压力而提出的。在P2P-VOD系统中视频的连续播放,服务器的被请求频率和用户的满意程度是系统重要因素,然而现在系统中的SVC缓存策略没有充分考虑到冷热门频道节点重叠存储数据的情况,由此本文提出MVC缓存策略,该策略充分考虑到点播系统中视频冷热度,流行值等会引起系统整体缓存不均衡的因素,提高冷门视频数据分块缓存量,在节点多的情况下,热门频道节点帮助冷门频道下载数据分片,在服务器通信量不足的情况下,提高视频播放的流畅程度,减轻服务器的负载压力。
本论文采用Oversim+INET+OMNeT++的方式在P2P网络上对该策略进行仿真时延。并从用户播放流畅度,服务器负载情况等性能上的仿真数据和曲线上,验证了该策略对SVC策略的改进。从仿真结果可以看出,该策略在一定程度上改善了SVC策略中,用户播放不流畅,频繁请求服务器的情况,具有较强的自适应性和实用性。
Data cache strategy is one of the key technologies of the streaming media software play quality, the usual cache strategy of the current streaming media player software is SVC (Single video cache), the strategy cache the data which peers are watching, However, in the current video-on-demand system, the user viewing behavior like Zipf, most users in the system demand the most popular video, but there are also exist a large number of cold video, because SVC stragety cache the current video user're watching, then in the network it caches more popular video data copy but less cold video video data, when peers watching the cold video, because there are less data in the network, it will exists the video on the peer will play unsmoothly and frequently requests the server to get the data segment they need, it is because exists such problem, so this paper, presents a P2P-VOD system cache strategy: Multi Video Cache(MVC), the strategy prompt out based on analysis the SVC cache strategy,according to the characteristics of the p2p-vod on demand system, in order to improve the fluency of the video playback and the server load. In P2P-VOD system the continuous of video playback, the requested of server and the degree of customer satisfaction are important factors in the system, but the SVC cache strategy in current system did not fully take into account of peers help each other cache the cold-hot video data. so the paper prompt the MVC cache strategy, the strategy take fully account of the hot-cold and the popular value which cause the the whole unequilibria in on-demand video system, improve the data quantity of cold video, in the case of peers are growing more, peers watching hot video help peers wathing cold video cache data segment, when the server traffic inadequate, improve the fluecy of the video and raise the server's node.
In order to analyze the performance of MVC strategy, the thesis utilizes Oversim + INET + OMNeTpp to simulate this strategy in P2P network. The simulation results are analyzed to verify the performance of MVC according these bases: user's play fluency, the server's load. After that, a conclusion can be drawn that the MVC cache stragety works relatively better than SVC at the paly fluency and peers frequency request the server. it has strong self-adaptability and practicality.
本论文采用Oversim+INET+OMNeT++的方式在P2P网络上对该策略进行仿真时延。并从用户播放流畅度,服务器负载情况等性能上的仿真数据和曲线上,验证了该策略对SVC策略的改进。从仿真结果可以看出,该策略在一定程度上改善了SVC策略中,用户播放不流畅,频繁请求服务器的情况,具有较强的自适应性和实用性。
Data cache strategy is one of the key technologies of the streaming media software play quality, the usual cache strategy of the current streaming media player software is SVC (Single video cache), the strategy cache the data which peers are watching, However, in the current video-on-demand system, the user viewing behavior like Zipf, most users in the system demand the most popular video, but there are also exist a large number of cold video, because SVC stragety cache the current video user're watching, then in the network it caches more popular video data copy but less cold video video data, when peers watching the cold video, because there are less data in the network, it will exists the video on the peer will play unsmoothly and frequently requests the server to get the data segment they need, it is because exists such problem, so this paper, presents a P2P-VOD system cache strategy: Multi Video Cache(MVC), the strategy prompt out based on analysis the SVC cache strategy,according to the characteristics of the p2p-vod on demand system, in order to improve the fluency of the video playback and the server load. In P2P-VOD system the continuous of video playback, the requested of server and the degree of customer satisfaction are important factors in the system, but the SVC cache strategy in current system did not fully take into account of peers help each other cache the cold-hot video data. so the paper prompt the MVC cache strategy, the strategy take fully account of the hot-cold and the popular value which cause the the whole unequilibria in on-demand video system, improve the data quantity of cold video, in the case of peers are growing more, peers watching hot video help peers wathing cold video cache data segment, when the server traffic inadequate, improve the fluecy of the video and raise the server's node.
In order to analyze the performance of MVC strategy, the thesis utilizes Oversim + INET + OMNeTpp to simulate this strategy in P2P network. The simulation results are analyzed to verify the performance of MVC according these bases: user's play fluency, the server's load. After that, a conclusion can be drawn that the MVC cache stragety works relatively better than SVC at the paly fluency and peers frequency request the server. it has strong self-adaptability and practicality.
引文
[1]李代玲.P2P流媒体技术[J].计算机与网络,2005年
[2]朱子荣.P2P点播系统媒体缓冲区研究与实现[硕士学位论文].合肥工业大学,2007年
[3]段翰聪.P2P流媒体内容分发技术研究[硕士学位论文].电子科技大学,2007年
[4]刘亚杰.P2P流媒体内容分发关键技术研究[硕士学位论文].国防科学技术大学研究生院,2005年9月
[5] cache.http://www.cachelogic.com
[6] Emule. http://www.emule.com
[7] BitTorrent. http://bitconjurer.org/BitTorrent/.
[8] PPLive. http://www.PPLive.com
[9] Xinyan Zhang, Jiangchuan Liu, Bo Li. CoolStreaming/DONet: A Data-Driven Overlay Network for Efficient Live Media Streaming[C]. Proceedings of IEEE INFOCOM, 2005: 2-5.
[10]视频点播系统介绍. http://hi.baidu.com/dorymei/blog/item/62672bfd04f1c740d7887dc6.html
[11]探讨IPTV媒体分发之视频流代理技术. http://www.cctime.com/html/2008-3-8/2008381100024017.htm.
[12] PPStream. http://www.PPStream.com
[13] The true picture of peer-to-peer file sharing [EB/OL].
[14]龚海刚,刘明,毛莺池,等.P2P流媒体关键技术的研究进展[J].计算机研究与发展,2005,42(12):2033-2040
[15]吴松,王浩,程斌,等. P2P VOD中基于多频道重叠网的数据缓存策略研究.小型微型计算机系统,2006.6
[16] Hou Meng-shu, Lu Xian-liang, Zhou Xu, et a1. Study on replication in unstructured P2P system[J]. Mini-Micro Systems, 2005, 26(11):1903-1906.
[17] Jin, Bin Cheng. GridCast: providing peer-to-peer on-demandstreaming service based on unstructured overlay[Z]. Submittedto the Workshop INPS 06
[18]董慧颖,陆莹,王洪斌,等.客户端缓存模拟器的设计与实现.信息技术,2008年第7期
[19]黄晓涛,郑涛.P2P流媒体点播的缓存机制研究.计算机工程与科学[J],2008年第30卷第3期
[20]廖小飞,殷江培,程斌.基于P2P的VOD系统中数据缓存策略研究[J].华中科技大学学报,2007年第35卷第8期
[21]李旭峰,陈海涛,王正华,等.通用P2P模拟器的构造技术研究.计算机应用研究,2006 Vol.23 No.5 P.16-18,37
[22]刘永卫,唐新春,刘戊开,等.基于缓存区段的P2P流媒体调度算法.计算机工程与科学,2008年第30卷第6期
[23] Yan Huang, Tom Z.J.Fu, Dah-Ming. Challenges, Design and Analysis of a Larte-scale P2P-VoD System[C]. Proceedings of ACM SIGCOMM[C], Settle, Washington, USA. August 2008 Page(s): 65-79.
[24]朱承,刘忠,张维明等.结构化P2P网络中基于流言传播的负载均衡[[J],通信学报.2004,第25卷第4期:31-40
[25]计算机世界报.2006年09月04日第34期B14、B15
[26] STOICAI, MORRIS R, KARGER D, et al. Chord: a scalable peer-to-peer lookup service for Internet applications [C], Proceedings of the International Conference of the Special Interest Group on Data Communication (SIGCOMM '01), Aug 27-31,2001, San Diego, CA, USA. New York, NY, USA: ACM Press, 2001:149-160
[27]杨天路,刘宇宏,张文等.P2P网络技术原理与系统开发实例,人民邮电出版社
[28]罗慧君.结构化P2P自适应高可用性负载均衡策略的研究与仿真[硕士学位论文].电子科技大学,2007年
[29]齐卫宁,王劲林.基于P2VoD的视频点播系统[J].计算机工程与应用,2008年44(10)
[30]缓存技术. http://www.cctime.com/html/2008-3-8/2008381100024017.htm
[31]流媒体. http://www.lmtw.com/sm/guandian/200505/7686.html
[32] VOD概念. http://www.cnvod.net/menu_vod.htm
[33]魏苏林,乐红兵.基于P2P模式的新型视频点播系统[J].微计算机信息,2005年第21卷第12-3期:1-2
[34]王文科.基于P2P的VoD系统技术研究与实现[硕士学位论文].太原理工大学,2007年
[35] Yang Guo, Kyoungwon Suh,Jim Kurose, et al.P2Cast:Peer-to peer Patching Scheme for VoD Service, National Science Foundation under NFS,May 20-24,2003
[36] X. Jiang, Y. Dong, D. Xu, et al. GnuStream: A P2P media streaming system prototype. In: Proc. 4th ICME’03. Los Alamitos, California: IEEE Computer Society Press, 2003
[37] Jin,Bin Cheng.GridCast:providing peer-to-peer on-demand streaming service based on unstructured overlay[Z].Submitted to the Workshop INPS 06.
[38] Zhenhua Li, Jiannong Cao, Guihai Chen. ContinuStreaming: Achieving High Playback Continuity of Goosip-based Peer-to_Peer Streaming[C]. IEEE, 2008.
[39] Jiangchuan Liu, Sanjay G.Rao, Bo Li. Opportunities and Challenges of Peer-to-Peer Internet Video Broadcase[C]. Proceedings of the IEEE, 2008
[40] Mohit Saxena, Umang Sharan, Sonia Fahmy. Analyzing Video Services in Web 2.0: A Global Perspective[C]. NOSSDAV, Braunschweig, Germany, 2008
[41] Razib lqbal, Behnoosh Hariri, Shervin Shirmohammadi. Modeling and Evaluation of Overlay Generation Problem for Peer-assisted Video Adaptation and Streaming. NOSSDAV, Braunschweig, Germany, 2008
[42] http://www.pcppc.cn/zuwang/guanliyuan/zuwang_81623_2.html
[43]李旭峰,陈海涛,王正华,等.通用P2P模拟器的构造技术研究.计算机应用研究,2006 Vol.23 No.5 P.16-18,37
[44] A. Varga. OMNeT++ community site. http://www.omnetpp.org/
[45]张小燕,吴辰文,葛立明.基于OMNeT++的IPv6协议仿真模型.微计算机信息,2007年,第03期:150-151,173
[46] A.Varga. OMNeT++ Object-oriented Discrete Event Simulation System User Manual.URL reference: http://www.omnetpp.org/external/doc/html/usman.php Jan 30,2004
[47] A.Varga OMNeT++ discrete event simulation system. In Proceedings of the European Simulation Multiconference(ESM’2001)
[48] Suzuki.J, Yamamoto.Y. iNet: an extensible framework for simulating immune network. Systems. Man and Cybernetics,2000IEEE International Conference on Volume 1:8-11 Oct.2000 Page(s):119-124 vol.1
[49] Calvert K., Doar M., Zegura E. Modeling Internet topology. IEEE Communication Magazine, 1997,vol.36,no.6:160-163
[50] Jared Winick, Sugih Jamin. Inet-3.0: Internet topology generator. Technical Report. CSE-TR-456-02, Ann Arbor: University of Michigan, 2002.
[51] Faloutsos M, Faloutsos P, Faloutsos C. On power-law relationships of the Internet topology. ACM SIGCOMM Computer Communication Review, 1999, 29(4):251-262.
[52] A.Huberman.Search in power law networks. Phys Rev46135-46143.
[53] I. Baumgart, B. Heep, and S. Krause. OverSim: A flexible overlay network simulation framework. Proceedings of 10th IEEE Global Internet Symposium(GI’07)in conjunction with IEEE INFOCOM 2007,Anchorage,AK,USA,May 2007
[54] Baumgar Ingmar, Hiip Bernhard, Krause Stephan. A P2P Demonstrator Powered by OverSim. Peer-to-Peer Computing, 2007.P2P 2007. Seventh IEEE International Conference on 2-5 Sept. 2007 Page(s):243-244
[2]朱子荣.P2P点播系统媒体缓冲区研究与实现[硕士学位论文].合肥工业大学,2007年
[3]段翰聪.P2P流媒体内容分发技术研究[硕士学位论文].电子科技大学,2007年
[4]刘亚杰.P2P流媒体内容分发关键技术研究[硕士学位论文].国防科学技术大学研究生院,2005年9月
[5] cache.http://www.cachelogic.com
[6] Emule. http://www.emule.com
[7] BitTorrent. http://bitconjurer.org/BitTorrent/.
[8] PPLive. http://www.PPLive.com
[9] Xinyan Zhang, Jiangchuan Liu, Bo Li. CoolStreaming/DONet: A Data-Driven Overlay Network for Efficient Live Media Streaming[C]. Proceedings of IEEE INFOCOM, 2005: 2-5.
[10]视频点播系统介绍. http://hi.baidu.com/dorymei/blog/item/62672bfd04f1c740d7887dc6.html
[11]探讨IPTV媒体分发之视频流代理技术. http://www.cctime.com/html/2008-3-8/2008381100024017.htm.
[12] PPStream. http://www.PPStream.com
[13] The true picture of peer-to-peer file sharing [EB/OL].
[14]龚海刚,刘明,毛莺池,等.P2P流媒体关键技术的研究进展[J].计算机研究与发展,2005,42(12):2033-2040
[15]吴松,王浩,程斌,等. P2P VOD中基于多频道重叠网的数据缓存策略研究.小型微型计算机系统,2006.6
[16] Hou Meng-shu, Lu Xian-liang, Zhou Xu, et a1. Study on replication in unstructured P2P system[J]. Mini-Micro Systems, 2005, 26(11):1903-1906.
[17] Jin, Bin Cheng. GridCast: providing peer-to-peer on-demandstreaming service based on unstructured overlay[Z]. Submittedto the Workshop INPS 06
[18]董慧颖,陆莹,王洪斌,等.客户端缓存模拟器的设计与实现.信息技术,2008年第7期
[19]黄晓涛,郑涛.P2P流媒体点播的缓存机制研究.计算机工程与科学[J],2008年第30卷第3期
[20]廖小飞,殷江培,程斌.基于P2P的VOD系统中数据缓存策略研究[J].华中科技大学学报,2007年第35卷第8期
[21]李旭峰,陈海涛,王正华,等.通用P2P模拟器的构造技术研究.计算机应用研究,2006 Vol.23 No.5 P.16-18,37
[22]刘永卫,唐新春,刘戊开,等.基于缓存区段的P2P流媒体调度算法.计算机工程与科学,2008年第30卷第6期
[23] Yan Huang, Tom Z.J.Fu, Dah-Ming. Challenges, Design and Analysis of a Larte-scale P2P-VoD System[C]. Proceedings of ACM SIGCOMM[C], Settle, Washington, USA. August 2008 Page(s): 65-79.
[24]朱承,刘忠,张维明等.结构化P2P网络中基于流言传播的负载均衡[[J],通信学报.2004,第25卷第4期:31-40
[25]计算机世界报.2006年09月04日第34期B14、B15
[26] STOICAI, MORRIS R, KARGER D, et al. Chord: a scalable peer-to-peer lookup service for Internet applications [C], Proceedings of the International Conference of the Special Interest Group on Data Communication (SIGCOMM '01), Aug 27-31,2001, San Diego, CA, USA. New York, NY, USA: ACM Press, 2001:149-160
[27]杨天路,刘宇宏,张文等.P2P网络技术原理与系统开发实例,人民邮电出版社
[28]罗慧君.结构化P2P自适应高可用性负载均衡策略的研究与仿真[硕士学位论文].电子科技大学,2007年
[29]齐卫宁,王劲林.基于P2VoD的视频点播系统[J].计算机工程与应用,2008年44(10)
[30]缓存技术. http://www.cctime.com/html/2008-3-8/2008381100024017.htm
[31]流媒体. http://www.lmtw.com/sm/guandian/200505/7686.html
[32] VOD概念. http://www.cnvod.net/menu_vod.htm
[33]魏苏林,乐红兵.基于P2P模式的新型视频点播系统[J].微计算机信息,2005年第21卷第12-3期:1-2
[34]王文科.基于P2P的VoD系统技术研究与实现[硕士学位论文].太原理工大学,2007年
[35] Yang Guo, Kyoungwon Suh,Jim Kurose, et al.P2Cast:Peer-to peer Patching Scheme for VoD Service, National Science Foundation under NFS,May 20-24,2003
[36] X. Jiang, Y. Dong, D. Xu, et al. GnuStream: A P2P media streaming system prototype. In: Proc. 4th ICME’03. Los Alamitos, California: IEEE Computer Society Press, 2003
[37] Jin,Bin Cheng.GridCast:providing peer-to-peer on-demand streaming service based on unstructured overlay[Z].Submitted to the Workshop INPS 06.
[38] Zhenhua Li, Jiannong Cao, Guihai Chen. ContinuStreaming: Achieving High Playback Continuity of Goosip-based Peer-to_Peer Streaming[C]. IEEE, 2008.
[39] Jiangchuan Liu, Sanjay G.Rao, Bo Li. Opportunities and Challenges of Peer-to-Peer Internet Video Broadcase[C]. Proceedings of the IEEE, 2008
[40] Mohit Saxena, Umang Sharan, Sonia Fahmy. Analyzing Video Services in Web 2.0: A Global Perspective[C]. NOSSDAV, Braunschweig, Germany, 2008
[41] Razib lqbal, Behnoosh Hariri, Shervin Shirmohammadi. Modeling and Evaluation of Overlay Generation Problem for Peer-assisted Video Adaptation and Streaming. NOSSDAV, Braunschweig, Germany, 2008
[42] http://www.pcppc.cn/zuwang/guanliyuan/zuwang_81623_2.html
[43]李旭峰,陈海涛,王正华,等.通用P2P模拟器的构造技术研究.计算机应用研究,2006 Vol.23 No.5 P.16-18,37
[44] A. Varga. OMNeT++ community site. http://www.omnetpp.org/
[45]张小燕,吴辰文,葛立明.基于OMNeT++的IPv6协议仿真模型.微计算机信息,2007年,第03期:150-151,173
[46] A.Varga. OMNeT++ Object-oriented Discrete Event Simulation System User Manual.URL reference: http://www.omnetpp.org/external/doc/html/usman.php Jan 30,2004
[47] A.Varga OMNeT++ discrete event simulation system. In Proceedings of the European Simulation Multiconference(ESM’2001)
[48] Suzuki.J, Yamamoto.Y. iNet: an extensible framework for simulating immune network. Systems. Man and Cybernetics,2000IEEE International Conference on Volume 1:8-11 Oct.2000 Page(s):119-124 vol.1
[49] Calvert K., Doar M., Zegura E. Modeling Internet topology. IEEE Communication Magazine, 1997,vol.36,no.6:160-163
[50] Jared Winick, Sugih Jamin. Inet-3.0: Internet topology generator. Technical Report. CSE-TR-456-02, Ann Arbor: University of Michigan, 2002.
[51] Faloutsos M, Faloutsos P, Faloutsos C. On power-law relationships of the Internet topology. ACM SIGCOMM Computer Communication Review, 1999, 29(4):251-262.
[52] A.Huberman.Search in power law networks. Phys Rev46135-46143.
[53] I. Baumgart, B. Heep, and S. Krause. OverSim: A flexible overlay network simulation framework. Proceedings of 10th IEEE Global Internet Symposium(GI’07)in conjunction with IEEE INFOCOM 2007,Anchorage,AK,USA,May 2007
[54] Baumgar Ingmar, Hiip Bernhard, Krause Stephan. A P2P Demonstrator Powered by OverSim. Peer-to-Peer Computing, 2007.P2P 2007. Seventh IEEE International Conference on 2-5 Sept. 2007 Page(s):243-244