近十幾年來隨著網路使用者與相關應用服務的倍增，同時也額外造成了許多軟硬體設備與網路頻寬的需求。為了對應這些需求與減少營運成本，Peer-to-Peer（P2P）的資料分享問題成為了一項熱門的研究議題。而其中如何在不違反網路上傳頻寬容量約束下，經由P2P檔案傳遞與任務排程將一組資料由單一個來源端，迅速送到多個接收端，更是P2P網路技術中重要的探討議題。這些研究除了探討如何改善資料完成傳遞的平均時間、避免重複傳送以及搜尋資料方法。

本論文研究的重點著重在改善P2P系統的資料派送，當P2P系統是由N個端點構成時，如何在2N個組合中找到P2P模式網路的最佳的資源排程，是屬於組合最佳化的問題，已被證明是屬於NP-Complete的問題，許多學者也嚐試以各種演算法用在解決該類問題上。本研究主要是採用結構式覆蓋網路（Structured Overlay Network）上傳頻寬管理方式，搭配G. Matthew Ezovski所提的Min-Min演算法為基礎提出兩項新的架構，第一種我們稱之為反向存儲和轉發（Reversed Store-and-Forward, RSF），這架構包含最長處理時間序優先排程（Longest Processing Time, LPT）與 Min-Min 演算法，其中提出幾項策略，包括初始化、區分工作項目、以改善P2P檔案傳遞頻寬容量排程。另一種我們稱之為分段式數據复用排程（Segmented Data Multiplexing, SDM），這架構包含區塊化資料和更新追蹤矩陣為基礎的資料架構。該區塊化資料使用固定數量與大小的資料區塊，並搭配二進位數列組成的ID與XOR機制來追蹤資料傳遞情況，然後以此為基礎與RSF共同構建一個結構式覆蓋網路以避免資料重複傳送的問題。

經由實驗結果顯示本研究所提出之方法有優於其他學者所提之Min-Min資料排程與一般BT檔案分享工具，在隨著使用者增加時可以明顯降低P2P檔案傳遞的平均結束時間。本研究還探討我們所提出P2P解決方案的可擴展性限制（scalability wall）與相對應的解決方案，藉此進一步改善使用者下載檔案的平均完成時間。

Peer-to-Peer (P2P) overlay networking is an important development for distributed systems and the evolution of Internet architecture. It has demonstrated its feasibility and economic potential for services involving millions of users. A fundamental challenge is to minimize average finish time, TA, of completing data delivery from a single source to multiple receivers with a constrained capacity for propagating data. When a network has n users and heterogeneous capacity, we have 2n possible deployments in the network, and the problem of finding out the best one has been proved to be NP-Complete. A great deal of research has followed to find the best deployments, thus we can minimize the empirical result of delay or finish time. Among many factors to affect TA, the critical considerations are capacity scheduling, data propagating and file locating.

The optimal timing reference of TA remains an open problem. This dissertation proposes two method of P2P file delivery that significantly reduces TA with understanding that peers often leave the network when completing file downloading. The methods construct structured overlay networks with an improved Min-Min algorithm for scheduling, and a data segment matrix for tracking and avoiding duplicate data delivery.

Two P2P test-beds consisting of hundreds of nodes with automatically generated queries are developed. Evaluation using these test-beds provides strong empirical evi-dences that the approaches proposed in this dissertation provide a better combination of efficiency and robustness than other common alternatives such as BT in peer-churn condition. Moreover, the results reveal that there can be a scalability wall of Amdahl's Law when increasing the number of peer in P2P overlay network. The P2P network can achieve superior performance by clustering the number of peer are in suitable size, and provide a substitute peer to replace the last peer.

摘要 iii
Abstract v
Content vii
List of Figures x
List of Tables xi
Chapter 1. Introduction 1
1.1 Background and Motivation 3
1.2 Challenges 5
1.2.1 Slow Server Condition 5
1.2.2 Limited Capacity 6
1.2.3 Data Lookup 6
1.2.4 Data Availability 7
1.2.5 Duplicate Transmission 8
1.3 Outline 9
Chapter 2. Related Works 10
2.1 Capacity Scheduling 12
2.2 Network Topology 14
2.2.1 Unstructured 15
2.2.2 Structured 16
2.3 Data Lookup 18
2.3.1 Blind Search 18
2.3.2 Informed Search 20
Chapter 3. Problem Formulation 23
3.1 Scope and Assumption 24
3.2 Average Finish Time 25
3.3 Notations 26
Chapter 4. Proposed Methods 27
4.1 RSF Scheduling Method 28
4.1.1 The First Time Interval 30
4.1.2 On the Later Time Intervals 32
4.1.3 Peer Remaining Support 33
4.2 SDM Scheduling Method 35
4.2.1 Divide Peers into Groups 36
4.2.2 Sequential Push-and-Pull of Data Segments 38
4.2.3 Data Segments 39
Chapter 5. Simulations and Analysis 42
5.1 RSF Scheduling Method 43
5.1.1 Impact of Peer-Leaving Condition 43
5.1.2 Impact of Different Capacity Distributions 45
5.1.3 Increasing the Size of Population with Peer-Churn 47
5.2 SDM Scheduling Method 52
5.2.1 The Finish Time of SDM 52
5.2.2 Comparison between SDM and BT 53
Chapter 6. Conclusions 58
6.1 Conclusions 58
6.2 Future Work 61
References 62
Appendix 71
A. Abbreviations 71

[1] Ipoque, (2010, February). "ipoque internet studies", [Online]. Available: http://www.ipoque.com/resources/internet-studies/internet-study-2008 2009
[2] Cisco, (2013, May 29) "Cisco Visual Networking Index: Forecast and Methodology 2012–2017", [Online]. Available:
http://www.cisco.com/en/US/solutions/collateral/ns341/ns525/ns537/ns705/ns827/white_paper_c11-481360_ns827_Networking_Solutions_White_Paper.html
[3] C. H. Wen, R. C. H. Lin, and S. W. Haga, " File Delivery with Longest Processing Time First Scheduling in P2P Networks", International Journal of Grid and Distributed Computing, vol. 6, no. 2, April, 2013
[4] K. Etminani, M. Naghibzadeh, "A Min-Min Max-Min selective algorithm for grid task scheduling", in proceedings of Internet, 3rd IEEE/IFIP International Conference in Central Asia on 2007, pp.1-7, 2007
[5] C. H. Wen and R. C. H. Lin, "Trim Down Duplicate transmission in Data Synchronization, Applied Mechanics and Materials, vol. 284, pp.2789-2793, 2013
[6] Y. Chawathe, S. Ratnasamy, and L. Breslau, “Making gnutellalike p2p systems scalable,” in Proceedings of SIGCOMM, Karlsruhe, Germany, August 2003, pp. 407–418.
[7] K. Ebcioglu, V. Saraswat, and V. Sarkar,"X10: Programming for hierarchical parallelism and non-uniform data access", in Proceedings of the 3rd International Workshop on Language Runtimes, 2004
[8] K. Yoshinaga, Y. Tsujita, A. Hori, M. Atsushi, M. Namiki, and Y. Ishikawa, Yutaka, "A Delegation Mechanism on Many-Core Oriented Hybrid Parallel Computers for Scalability of Communicators and Communications in MPI", in Proceedings of International Conference on Parallel, Distributed and Network-Based Processing (PDP), IEEE, pp.249-253, 2013
[9] G. M. Ezovski, A. Tang, and L. Andrew, "Minimizing average finish time in p2p networks", in Proceedings of IEEE International Conference on Computer Communications (INFOCOM), pp. 594-602, 2009.
[10] X. Shen, H. Yu, J. Buford, and M. Akon, "Handbook of peer-to-peer networking", Springer Publishing Company, 2009.
[11] P. Brucker, "Scheduling algorithms", Springer Verlag, 5th ed., 2007.
[12] S. Deb, A. Ganesh and P. Key, "Resource allocation between persistent and transient flows", IEEE/ACM Transactions on Networking (TON), vol.13, no.2, pp.302-315, 2005.
[13] G. Huang and Y. Zhou, "A Request Handling Mechanism with the Shortest Processing-Time First in Web Servers", in Proceedings of IEEE Computer Science and Software Engineering (CSSE), vol. 3, 2008.
[14] G. Wei, Y. Ling, Y. Gu, and Y. Ge, "A dependable cluster based topology in p2p networks", Journal of Communications, vol. 5, no. 1, pp. 57–62, 2010.
[15] D. Stutzbach and R. Rejaie, "Understanding churn in peer-to-peer networks", in Proceedings of ACM SIGCOMM on Internet measurement, ACM, 2006, p. 202.
[16] M. Mehyar, W. Gu, S. Low, M. Effros, and T. Ho, "Optimal Strategies for Efficient Peer-to-Peer File Sharing," in Proceedings of IEEE ICASSP, 2007.
[17] E. K. Lua, J. Crowcroft, M. Pias, R. Sharma and S. Lim, "A Survey and Comparison of Peer-to-Peer Overlay Network Methods", IEEE Communications Surveys and Tutorials, Second Quarter 2005, Volume 7, No. 2., pp.72-93
[18] Z. Wang, C. Wu, L. Sun, and S. Yang, "Strategies of collaboration in multi-swarm peer-to-peer content distribution", Tsinghua Science and Technology, vol.17, no.1, pp.29-39, 2012
[19] S. Androutsellis-Theotokis, and D. Spinellis, "A survey of peer-to-peer content distribution technologies", ACM Computing Surveys , vol.36, no.4, pp.335-371, 2004
[20] Y. Huang, T.Z.J Fu, Chiu D, J. Lui, and C. Huang, "Challenges, design and analysis of a large-scale p2p-vod system", ACM SIGCOMM Computer Communication Review, vol.38, no.4, pp.375-388, 2008
[21] J.Y. Chuang, P.L. Hsu, and Y.S. Chen, "Improving Broadcast Processes in Peer-to-Peer Systems (Invited Extended Paper)", in Proceedings of The 7th International Conference on Networked Computing, pp.48-53, 2011
[22] R.A. Ferreira, M.K. Ramanathan, A. Grama, and S. Jagannathan, "Randomized protocols for duplicate elimination in peer-to-peer storage systems", IEEE Transactions on Parallel and Distributed Systems, IEEE, vol.18, no.5, pp.686-696, 2007
[23] B. Fan, J. Lui, and D. Chiu, "The design trade-offs of BitTorrent-like file sharing protocols", IEEE/ACM Transactions on Networking (TON), IEEE/ACM, vol. 17, no. 2, pp. 365–376, 2009.
[24] M. Piatek, T. Isdal, T. Anderson, A. Krishnamurthy, and A.Venkataramani, "Do incentives build robustness in bittorrent?", in Proceedings of Networked Systems Design and Implementation (NSDI), vol. 7, 2007.
[25] Y. Tian, D. Wu, and K. Ng, "Modeling, analysis and improvement for bittorrent-like file sharing networks", in Proceedings of IEEE International Conference on Computer Communications (INFOCOM), IEEE, 2006, pp. 1–11.
[26] F. B. Y. Qiao, D. Lu, "Improving peer-to-peer performance through server-side scheduling", ACM Transactions on Computer Systems, ACM, vol. 26, Dec. 2008.
[27] D. Qiu and R. Srikant, "Modeling and performance analysis of bittorrent-like peer-to-peer networks", ACM SIGCOMM Computer Communication Review, ACM, vol. 34, no. 4, pp. 367–378, 2004.
[28] R. Kumar and K. Ross, "Peer assisted file distribution: The minimum distribution time", in Proceedings of First IEEE Workshop on Hot Topics in Web Systems and Technologies (HOTWEB), IEEE, 2006.
[29] Y. Lu, A. Zhang, H. He, and Z. Deng, "Stochastic fluid model for P2P content distribution networks", in Proceedings of Autonomous Decentralized Systems (ISADS), IEEE , pp. 707-712, 2005
[30] X. Yang and G. De Veciana, “Service capacity of peer to peer networks,” in Proceedings of IEEE International Conference on Computer Communications (INFOCOM), IEEE, vol. 4, 2004.
[31] R. R. W. J. Mundinger and G. Weiss, "Analysis of peer-to-peer file dissemination amongst users of different upload capacities", ACM SIGMETRICS Performance Evaluation Review, vol. 34, no. 2, Sept. 2006, pp. 5–6.
[32] Graham, R.L. and Lawler, E.L. and Lenstra, J.K. and Kan, A.H.G.R., "Optimization and approximation in deterministic sequencing and scheduling: A survey", Annals of Discrete Mathematics, vol. 5, no. 2, pp. 287–326, 1979.
[33] J. Ullman, "NP-complete scheduling problems", Journal of Computer and System Sciences, vol. 10, no. 3, pp. 384–393, 1975.
[34] R.T. Ma, S. Lee, J. Lui, and D.K. Yau, "Incentive and service differentiation in P2P networks: a game theoretic approach", IEEE/ACM Transactions on Networking (TON), vol.14, no.5, pp.978-991, 2006
[35] C. Liaw, Y. Lin, C. Cheng, and M. Chen, "Scheduling unrelated parallel machines to minimize total weighted tardiness", Computers & Operations Research, vol. 30, no. 12, pp. 1777–1789, 2003.
[36] E. Modiano, “Random algorithms for scheduling multicast traffic in wdm broadcast-and-select networks,” IEEE/ACM transactions on Networking (TON), vol. 7, no. 3, pp. 425–434, 1999.
[37] L. Wang and D. Li, "A scheduling algorithm for flexible flow shop problem", in Proceedings of World Congress on Intelligent Control and Automation, vol. 4, 2002.
[38] M. van den Akker and H. Hoogeveen, "Minimizing the number of late jobs in a stochastic setting using a chance constraint", Journal of Scheduling, vol. 11, no. 1, pp. 59–69, 2008.
[39] K. Shibata and M. Yokoi, "Optimum scheduling in pipeline processing", in Proceedings of IEEE ICCS/ISITA ’Communications on the Move’, vol. 13, pp. 1078–1083, 1992.
[40] J. Mundinger and R. Weber, "Efficient file dissemination using peer-to-peer technology. technical report 2004", Statistical Laboratory, Cambridge Tech. Report, Jan. 2004.
[41] C. Wu, B. Li, S. Zhao, and I. UUSee, “Multi-channel live p2p streaming: Refocusing on servers,” in Proceedings of IEEE International Conference on Computer Communications (INFOCOM), IEEE, Apr. 2008.
[42] L. O. Alima, A. Ghodsi, and S. Haridi, "A framework for structured peer-to-peer overlay networks", Global computing, Lecture Notes in Computer Science: Springer Berlin / Heidelberg, vol. 3267, pp. 223-24, 2005.
[43] D. Liben-Nowell, H. Balakrishnan, and D. Karger, "Analysis of the Evolution of Peer-to-Peer Systems", Proceeding of PODC, 2002.
[44] M. Roussopoulos, M. Baker, D. Rosenthal, T. Guili, P. Maniatis, and J. Mogul, "2 P2P of Not 2 P2P?", in proceedings of The 3rd Int'l Workshop on Peer-to-Peer Systems, February 26-27 2004.
[45] A. Rowstron and P. Druschel, "Pastry: Scalable, distributed object location and routing for large-scale peer-to-peer systems", IFIP/ACM Middleware 2001, Nov 2001.
[46] B. Zhao, L. Huang, J. Stribling, S. Rhea, A. Joseph, and J. Kubiatowicz, "Tapestry: A Resilient Global-Scale overlay for Service Deployment", IEEE Journal on Selected Areas in Communications, vol. 22, no.1, pp.41-53, 2004
[47] R. Huebsch, J. M. Hellerstein, N. Lanham, B. T. Loo, S. Shenker, and I. Stoica, "Querying the Internet with PIER", in Proceedings of 29th Int'l Conference on Very Large Databases VLDB'03, September 2003.
[48] J. M. Hellerstein, "Toward network data independence", in Proceedings of ACM SIGMOD Record, vol.32, no.3, 2003, 34-40.
[49] R. Rodrigues and P. Druschel, "Peer-to-Peer systems", Communications of the ACM, vol. 53, no. 10, pp. 72–82, 2010.
[50] B.Y. Zhao, L. Huang, J. Stribling, et al., "Tapestry: a resilient global-scale overlay for service deployment", IEEE Journal on Selected Areas in Communications, vol. 22, no. 1, pp. 41–53, 2004.
[51] Z. Ou, E. Harjula, O. Kassinen, et al., "Performance evaluation of a Kademlia-based communication-oriented P2P system under churn", Computer Networks, vol. 54, no. 5, pp. 689–705, 2010.
[52] I. Stoica, R. Morris, D.R. Karger, et al., "Chord: a scalable peer-to-peer lookup service for Internet applications", IEEE/ACM Transactions on Networking, vol. 11, no. 1, pp. 17–32, 2003.
[53] Y.J. Joung and J.C. Wang JC, “Chord2: a two-layer Chord for reducing maintenance overhead via heterogeneity,” Computer Networks, vol. 51, no. 3, pp. 712–731, 2007.
[54] S. Ratnasamy, P. Francis, M. Handley, R. Karp, and S. Shenker, "A scalable content-addressable network", in Proceeding of the conference on Applications, technologies, architectures and protocols for computer communications, pp. 161-172, August 2001.
[55] D. Malkhi, M. Naor, and D. Ratajczak, "Viceroy: a scalable and dynamic emulation of the butterfly", in Proceedings of 21st annual symposium on principles of distributed computing PODC, pp. 183-192, July 2002.
[56] R. A. Rowley and B. Bose, "Fault-tolerant ring embedding in de Bruijn networks, IEEE Transaction on Computers 42 (12) (1993) 1480-1486
[57] P. Fraigniaud and P. Gauron, "The content-addressable networks D2B", Technical Report 1349, Laboratoire de Recherche en Informatique, January 2003.
[58] F. Kaashoek and D. Karger, Koorde, "A simple degree-optimal hash table", in proceedings of Second Int'l Workshop on Peer-to-Peer Systems (IPTPS), February 2003.
[59] M. Naor and U. Wieder, "A Simple Fault Tolerant Distributed Hash Table", in proceedings of Second Int'l Workshop on Peer-to-Peer Systems (IPTPS), February 2003.
[60] D. Loguinov, A. Kumar, and S. Ganesh, "Graph-theoretic analysis of structured peer-to-peer systems: routing distances and fault resilience", in Proceedings of 2003 conference on Applications, Technologies, Architectures and Protocols for Computer Communications, pp. 395-406, August 2003.
[61] N. Harvey, M. B. Jones, S. Saroiu, M. Theimer, and A. Wolman, "SkipNet: A Scalable Overlay Network with Practical Locality Properties", in Proceedings of Fourth USENIX Symp. on Internet Technologies and Systems (USITS), March 2003.
[62] J. Cates, "Robust and Efficient Data Management for a Distributed Hash Table", Master's Thesis, May 2003.
[63] J. Mishchke and B. Stiller, "A methodology for the design of distributed search in P2P middleware", IEEE Network, IEEE, vol.18, no.1, pp.30-37, 2004.
[64] Y. Chawathe, S. Ratnasamy, L. Breslau, N. Lanham, and S. Shenker, "Making gnutella-like P2P systems scalable", in Proceedings of 2003 conference on Applications, Technologies, Architectures and Protocols for Computer Communications, pp. 407-418, August 2003.
[65] K. Gummadi, R. Gummadi, S. Gribble, S. Ratnasamy, S. Shenker, and I. Stoica, "The impact of DHT routing geometry on resilience and proximity", in Proceedings of 2003 conference on Applications, Technologies, Architectures and Protocols for Computer Communications, pp. 381-394, 2003.
[66] D. Loguinov, A. Kumar, and S. Ganesh, "Graph-theoretic analysis of structured peer-to-peer systems: routing distances and fault resilience", in Proceedings of 2003 conference on Applications, Technologies, Architectures and Protocols for Computer Communications, pp. 395-406, August 2003.
[67] M. Castro, M. Costa, and A. Rowstron, "Should we build Gnutella on a structured overlay?", ACM SIGCOMM Computer communication Review, ACM, vol.34, no.1, pp.131-136, 2004
[68] B. T. Loo, R. Huebsch, I. Stoica, and J. M. Hellerstein, "The case for a hybrid P2P search infrastructure", in proceedings of The 3rd Int'l Workshop on Peer-to-Peer Systems, pp.26-27, 2004.
[69] C. Gkantsidis, M. Mihail, and A. Saberi, “Random Walks in Peer-to-Peer Networks,” in Proceedings of IEEE INFOCOM, IEEE, vol.1, 2004.
[70] C. Lv, P. Cao, E. Cohen, K. Li, and S. Shenker, "Search and replication in unstructured peer-to-peer networks", in Proceedings of ACM SIGMETRICS, ACM, 2002.
[71] L. A. Adamic, R. M. Lukose, A. R. Puniyani, and B. A. Huberman, "Search in power law networks", Physical Review E, vol.64, no.4, pp.46135-46143, 2001
[72] J. Liang, R. Kumar, and K. Ross, "The FastTrack overlay: A measurement study", Computer Networks, Elsevier, vol.50, no.6, pp.842-858, 2006
[73] H. Unger and M. Wulff, "Cluster-building in p2p-community networks", Parallel and Distributed Computing and Systems, pp. 685–690, 2002
[74] C. Lv, P. Cao, E. Cohen, K. Li, and S. Shenker, "Search and replication in unstructured peer-to-peer networks", in Proceedings of the 16th international conference on Supercomputing, pp.84-95, 2002.
[75] S. Joseph, " Neurogrid: Semantically routing queries in peer-to-peer networks", Web Engineering and Peer-to-Peer Computing, Springer, pp.202-214, 2002.
[76] A. Crespo and H. Garcia-Molina, "Routing indices for peer-to-peer systems", in Proceedings of 22nd International Conference on Distributed Computing Systems, pp. 23-33, 2002.
[77] L. Xu, C. Dai, W. Cai, S. Zhou, and A. Zhou,"Towards adaptive probabilistic search in unstructured P2P systems", Advanced Web Technologies and Applications, Springer, pp. 258-268, 2004.
[78] Guofu, Feng and Yingchi, Mao and Sanglu, Lu and Daoxu, Chen, "Small World Based Adaptive Probabilistic Search (SWAPS) for Unstructured Peer-to-Peer File Systems", Journal of Computer Research and Development, vol.3, p3, 2006.
[79] D. England, B. Veeravalli, and J. Weissman, "A Robust Spanning Tree Topology for Data Collection and Dissemination in Distributed Environments", IEEE Transactions On Parallel and Distributed Systems, vol. 18, no. 5, pp. 608–620, 2007.
[80] A. Kumar, S. Merugu, J. Xu, and X. Yu: Ulysses, "A Robust, Low-Diameter, Low-Latency Peer-to-Peer Network", European Transaction on Telecommunications, vol. 15, no. 6, pp. 571–587, 2004.
[81] J. S. Li and C. H. Chao,"An efficient superpeer overlay construction and broadcasting method based on perfect difference graph", IEEE Transactions on Parallel and Distributed Systems, vol. 21, no. 5, pp. 594-606, 2010
[82] T. Nguyen, S. Cheung, and D. Tran, "Efficient p2p file delivery in a homogeneous capacity network using structured mesh," in IEEE International conference on multimedia access networks, 2005.
[83] S. Ratnasamy, S. Shenker, and I. Stoica, "Routing algorithms for DHTs: Some open questions", Peer-to-Peer Systems, Springer, pp.45-52, 2002
[84] H. Shen and C. Z. Xu, "Hash-based proximity clustering for efficient load balancing in heterogeneous DHT networks", Journal of Parallel and Distributed Computing, Elsevier, vol.68, no.5, pp.686-702, 2008
[85] Microsoft, "Microsoft research simulator for the bittorrent protocol", [Online] Available: http://research.microsoft.com/en-us/downloads/20d68689-9a8d-44c0-80cd-66dfa4b0504b/
[86] A. Farzad and H. R. Rabiee, "Modeling Topological Characteristics of BitTorrent-Like Peer-to-Peer Networks", IEEE Communication Letters, vol. 15, no. 8, pp.896–898, 2011.
[87] Evans T.S. & Plato A.D.K., "Exact Solution for the Time Evolution of Network Rewiring Models", Physical Review E, Vol. 75, No. 5, 2007, p.56101.
[88] Fro`nczak A. & Fronczak P. & Hołyst J., "Average Path Length in Random Networks", Physical Review E, Vol. 70, No. 5, 2004, p.56110.
[89] Menéndez-Leonel C.A. & Benítez-Pérez H., "Scheduling strategy for real-time distributed systems", Journal of Applied Research and Technology, Vol. 8, No. 2, 2010, pp.177-184.