臺灣博碩士論文加值系統

對等式(Peer-to-peer)網路應用程式在近年來引起很大的注意。對等式系統和傳統的大型主機系統(client-server)是非常不一樣的，在對等式系統裡，每一在系統中的peer都可以和其他的peer共享資源。
在對等式系統中的每個 peer可以根據其需求在伺服器端或客戶端之間做轉換。如果某一個特定的peer當機或離線，其他擁有同樣資源的peer可以取而代之繼續提供相同的服務。在這一篇論文中，我們在對等式系統上建立一個網頁伺服器，這個系統，稱為Sniper，可支援高度獲得(highly-available)，負載分享，使用者可自行管理等特性。
在Sniper系統中，每一個peer可經由網際網路來共享他們的磁碟空間，而且沒有所謂的系統管理者存在於此系統中。每一份網頁的發行者同時也是網頁的管理者。我們在論文中會詳細地描述在Sniper系統中使用者的要求及備份如何達成。我們同時也提出一個chain replication algorithm來達到高可靠度。經由這個演算法，我們可以在Sniper系統中對於每份發行的網頁維持K個備份，我們同時也修改了routing algorithm，使其能分散使用者的要求到存有備份的peer上，進而達到負載分享。

Peer-to-peer Internet applications have recently attracted much attention. Peer-to-peer systems are different from traditional server-client systems. In peer-to-peer systems, every peer shares its resources with other peers. A peer can be a server at some time and a client at other time. In this thesis, we build a web server on top of a peer-to-peer infrastructure. The resultant system, called Sniper, can support a highly-available, load-sharing, user-administrable web server.
In the Sniper system, every peer shares its disk space to form a large logical web space on the Internet. There is no system administrator in the system. Every content publisher is also the administrator of his own web contents. We describe how requests and replicas can be routed among the peers following a peer-to-peer paradigm. We propose a chain replica algorithm to achieve high availability. With the algorithm, the Sniper system can always maintain a constant number of replicas. We also extend the routing algorithm to distribute the requests to the replicas for load sharing. Performance of Sniper is finally evaluated through simulation.

第一章 簡介
第二章 相關研究工作
第三章 系統設計
第四章 複製管理
第五章 實驗及結果
第六章 結論及未來展望
英文附錄

[1] A. Rowstron, and P. Druschel, “Storage management and caching in PAST, a large-scale, persistent peer-to-peer storage utility”, Proc. ACM SOSP’01, Banff, Canada, Oct. 2001.
[2] Akamai. http://www.akamai.com/.
[3] A. Rowstron and P. Druschel, “Pastry: Scalable, decentralized object location and routing for large-scale peer-to-peer systems”, Proceeding of the 18th IFIP/ACM International Conference on Distributed Systems Platforms (Middleware 2001), Heidelberg, Germany November 2001.
[4] B. Y. Zhao, J. D. Kubiatowicz, and A. D. Joseph, “Tapestry: An infrastructure for fault-resilient wide-area location and routing”, Technical Report UCB//CSD-01-1141, U. C. Berkeley, April 2001.
[5] B. Nooble, B. Fleis, M. Kim, “A case for fluid replication”, Proceeding of the 1999 Network Storage Symposium (Netstore), Oct. 1999.
[6] Gnutella. http://www.gnutella.com/.
[7] H. Hsiao and C. King, “Torando: A scalable and fault-resilient peer-to-peer self-organizing infrastructure”, Technical Report, Department of Computer Science, National Tsing Hua University, 2002.
[8] I. Goldberg and D.Wagner, “TAZ servers and the rewebber network: Enabling anonymous publishing on the world wide web”. First Monday, 3, 1998.
[9] J. Kubiatowicz, D. Bindel, P. Eaton, Y. Chen, D. Geels, R. Gummadi, S. Rhea, W. Weimer, C. Wells, H. Weather-spoon, and B. Zhao, “OceanStore: An architecture for global-scale persistent store”, Proc. ACM ASPLOS’2000, Cambridge, MA, November 2000.
[10] H. T. Kung, and S. Bradner, “A Framework for an anonymizing packet forwarder”, Draft, November 2001.
[12] M. Castro, P. Druschel, A. Kermarrec and A. Rowstron, “SCRIBE: A large-scale and decentralized publish-subscribe infrastructure”, Proceedings of the Third International Workshop on Networked Group Communications (NGC2001). London, UK, November 2001.
[11] M. Waldman, A. D. Rubin, L. Cranor, “Publius: A robust, tamper-evident, censorship-resistant web publishing system”, Proceeding of the 9th USENIX Security Symposium (August 2000), pp.59-72.
[13] Napster. http://www.napster.com/.
[14] P. Druschel and A. Rowstron, "PAST: A large-scale, persistent peer-to-peer storage utility", Proc. HotOS VIII, Schoss Elmau, Germany, May 2001.
[15] SHA. http://philosophysw.com/.
[16] A. Schiper, J. Eggi, and A. Sandoz, “A new algorithm to implement causal ordering”, Proceeding of the International Workshop on Distributed Algorithms, Springer-Verlag, New York, 1989, pp.219-232.
[17] D. B. Terry, M. M. Theimer, K. Petersen, and A. J. Demers, “Managing update conflicts in Bayou, a weakly connected replicated storage system”, Proceeding of the Fifteenth ACM Symposium on Operating System Principles, p.172-83. Dec 1995, Copper Mountain, CO, USA.
[18] Y. Chen, R. H. Katz and J. D. Kubiatowicz, “Dynamic replica placement for scalable content delivery”, Proceeding of 1st International Workshop on Peer-to-Peer Systems (IPTPS'02), Boston, Mar. 2002.