點對點式(peer-to-peer, 簡稱P2P)應用的興起所產生的大量跨網域流量對網際網路服務提供者(Internet Service Provider, 簡稱ISP)的維運成本構成了可觀的威脅。其主要問題來自於上層覆蓋網路與底層實體網路的不對稱。最近，BitTorrent 以其大規模傳輸檔案的便利性吸引了大量的使用者，也變成了ISP的一個挑戰。因此，許多加強BitTorrent流量區域性的解決辦法被提出來。然而，有些解法不能單靠ISP來完成而必須對BitTorrent的運作機制加以修改。因為這個問題只對ISP有較為迫切的影響，造成了鮮少人採用這些解法的窘境。一個較適合ISP的方法可能是快取。但大部分基於快取的解決辦法通常只考慮運作在一個BitTorrent覆蓋網這樣不實際的環境下。所以，本篇論文提出了一個使用快取以區域化BitTorrent流量的機制，提供給ISP來減緩大量的跨網域流量。一些高資源使用效率的方法也被提出來以分配有限的資源於多個覆蓋網的流量區域化。本篇所提出B-Proxy的效用也經過在PlanetLab上的實際模擬來加以評估，每次實驗都包含上百個BitTorrent的客戶端軟體被執行。實驗結果顯示，對單一個覆蓋網來說，超過百分之三十的跨網域流量可以被節省卻只消耗了八分之一大小的檔案空間用以儲存快取。

The emergence of peer-to-peer (P2P) applications has posed a threat to the operating cost of Internet Service Providers (ISPs) due to the large amount of inter-ISP traffic generated. The problem stems from the mismatch between the P2P overlay network formed randomly and the underlying physical network. Recently, BitTorrent has attracted enormous users by its convenience of large-scale content distribution and has also become a major challenge for ISPs. Therefore, a great deal of study has been done to enhance the traffic locality in BitTorrent. However, some of them could not be achieved by ISPs alone and need to change the original mechanism of BitTorrent, making these solutions hardly used in practice since the problem is only urgent to ISPs. The most suitable solution for ISPs to deploy is probably cache. Nevertheless, most cache-based solutions only account for a single BitTorrent overlay (torrent) which is unpractical. Thus, this thesis presents a cache-based BitTorrent traffic localization scheme designed for ISPs to ease the volume of inter-ISP traffic generated by BitTorrent and also proposes several resource-efficient policies to allocate the limited resources for multiple torrents. The effectiveness of the proposed B-Proxy was also evaluated through realistic simulation on PlanetLab, where hundreds of BitTorrent clients were executed during the experiment. Simulation results show that more than thirty percent of inter-ISP traffic could be saved in a torrent with a relatively small cache size consumed which is only eighth times that of the original file.

[1] Akamai. http://www.akamai.com
[2] BitTorrent. Free, open source file-sharing application effective for distributing very large software and media files.
http://www.bittorrent.com
[3] The Pirate Bay, the world’s largest BitTorrent tracker.
http://thepiratebay.org/
[4] Ipoque. http://www.ipoque.com/
[5] P-Cube. http://www.p-cube.com
[6] Sandvine. http://www.sandvine.com/
[7] Ruchir Bindal, Pei Cao, William Chan, Jan Medved, George Suwala, Tony
Bates, and Amy Zhang, “Improving Traffic Locality in BitTorrent via Biased
Neighbor Selection,” in Proceedings of IEEE ICDCS’06, Lisboa, Portugal,
July 2006, pp. 66-66.
[8] Yunhao Liu, Xiaomei Liu, Li Xiao, Lionel M. Ni, and Xiaodong Zhang, “Location-aware topology matching in P2P systems,” in Proceedings of IEEE INFOCOM’04, Hong Kong, March 2004, pp.2220-2230 vol. 4.
[9] Yunhao Liu, Li Xiao, Xiaomei Liu, Lionel M. Ni, and Xiaodong Zhang, “Location awareness in unstructured peer-to-peer systems,” IEEE Transactions on Parallel and Distributed Systems, Vol. 16, Issue 2, 2005, pp. 163-174.
[10] BitTorrent clients.
http://en.wikipedia.org/wiki/BitTorrent_client
[11] CacheLogic – advanced solutions for peer-to-peer networks.
http://www.velocix.com/
[12] Adam Wierzbicki, Nathaniel Leibowitz Matei Ripeanu, and Rafal Wozniak, “Cache replacement policies revisited: the case of P2P traffic,” in Proceedings of the 4th IEEE/ACM International Symposium on Cluster Computing and the Grid(CCGrid’04), Chicago, Illinois, USA, April 2004, pp. 182-189.
[13] Osama Saleh and Mohamed Hefeeda, “Modeling and Caching of Peer-to-Peer Traffic,” in Proceedings of the 14th IEEE International Conference on Network Protocols (ICNP '06), Santa Barbara, California, USA, November, 2006, pp. 249-258.
[14] PlanetLab. http://www.planet-lab.org/
[15] Bram Cohen, “Incentives Build Robustness in BitTorrent,” in the 1st Workshop on Economics of Peer-to-Peer Systems, Berkeley, CA, USA, June 2003.
[16] Lei, Zhang, “Study of the location awareness in bitTorrent-like networks,” in the 7th International Conference on Computer-Aided Industrial Design and Conceptual Design (CAIDCD '06), HangZhou, CHINA, November 2006, pp.1-7.
[17] Thomas Karagiannis, Pablo Rodriguez, and Konstantina Papagiannaki, “Should internet service providers fear peer-assisted content distribution?” in Proceedings of the Internet Measurement Conference (IMC’05), Berkeley, CA, USA, October 2005, pp. 6-6.
[18] Nathaniel Leibowitz, Aviv Bergman, Roy Ben-Shaul, and Aviv Shavit, “Are file swapping networks cacheable? Characterizing P2P traffic,” in Proceedings of the 7th International Workshop on Web Content Caching and Distribution(WCW’02), Boulder, Colorado, USA, 2002.
[19] Mauro Andreolini, Riccardo Lancellotti, and Philip S. Yu, “Analysis of peer-to-peer systems: workload characterization and effects on traffic cacheability,” in Proceedings of the IEEE Computer Society's 12th Annual International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunications Systems (MASCOTS’04), Volendam, Netherlands, October 2004, pp.95-104.
[20] Cache Discovery Protocol.
http://en.wikipedia.org/wiki/Cache_Discovery_Protocol
[21] Thanunchai Threepak, “Bittorrent Cache Using Virtual Tracker,” in Proceedings of the IEEE International Symposium on Communications and Information Technologies (ISCIT '06), Bangkok, Thailand, October 2006, pp. 162-165.
[22] Enhanced Ctorrent. http://www.rahul.net/dholmes/ctorrent/
[23] Libpcap. http://www.tcpdump.org/
[24] National Ilan University. http://www.niu.edu.tw
[25] BitComet. http://www.bitcomet.com
[26] Xunlei. http://www.xunlei.com
[27] FlashGet. http://www.flashget.com
[28] PlanetLab experimental manager.
http://www.cs.washington.edu/research/networking/cplane/
[29] Ashwin R. Bharambe, Cormac Herley, and Venkata N. Padmanabhan, “Analyzing and Improving a BitTorrent Networks Performance Mechanisms,” in Proceedings of IEEE INFOCOM’06, Barcelona, Catalunya, Spain, April, 2006, pp. 1-12.