臺灣博碩士論文加值系統

內容傳遞網路(Content Delivery Network, CDN)，是將使用者的請求分散至各處的代理緩存伺服器(Surrogate server)，以減輕原始伺服器(Origin server)的負擔，並避開網路中的瓶頸（Bottleneck）和壅塞部分，增加傳輸速度以減少回應時間，進而提高整體網路的效能。CDN以重新導向讓使用者連線導向至較佳代理緩存伺服器，代理伺服器提供使用者能以較短時間取得內容，且能降低長距離傳輸內容而導致封包遺失、封包延遲的情況。但目前所提出的重新導向機制在設計上不夠完善，目前所提出的重新導向只能在使用者新建連線時才作重新導向，但可能因此導致新建連線時使用者的最佳代理緩存伺服器滿載，使得使用者被導向至次佳代理緩存伺服器；當原先滿載的代理緩存伺服器有空閒時，卻無法將較差代理緩存伺服器導向至較佳代理緩存伺服器。
本論文所設計的重新導向演算法使用內容管理器(Content manager)監控代理緩存伺服器的負載狀況，並且記錄沒能取得最佳代理緩存伺服器的使用者，有機會將使用者再次重新導向到較佳的代理緩存伺服器，使得能更即時取得內容。本論文透過模擬的結果，分析重新導向演算法為終端用戶所帶來的效益。

Content Delivery Networks (CDN) are composed of multiple surrogate servers located at different geographic locations which redirection user requests for the goal of mitigating the load of the origin Server, avoiding busy transmission paths, and increasing the data transfer rate to lower response time, in the end improving overall network performance. CDN uses methods of redirection that forward the user connections to the suitable surrogate servers that offer a lower response time for accessing data, in the meantime reducing the packet losses and delays possible in long transmission paths. However, the current mechanisms for redirection are flawed, as they can only perform redirection when a user connection is established. This may cause the target surrogate server to become overloaded during a newly created connection, as a result redirecting the user to a less suitable surrogate server. A previously overloaded surrogate server may also become idle, yet unable to redirect a user to a more suitable surrogate server from a less suitable surrogate server.
This thesis provides a redirection algorithm that uses a content manager to monitor the work load of surrogate servers, and maintains a record of the users that did not obtain the optimal suitable surrogate servers. This allows users to be redirected repeatedly until a more suitable surrogate server is found and allows a more instantaneous access of content. This thesis analyzes the performance improvements of terminal users applying the algorithm through simulation results.

摘　要 i
ABSTRACT ii
誌　謝 iv
目　錄 v
表目錄 vii
圖目錄 viii
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機 2
1.3 論文架構 2
第二章 相關技術背景之探討 3
2.1 CDN技術簡介 3
2.1.1 CDN組成 4
2.1.2 內容選擇與傳送 5
2.1.3 內容的放置 8
2.1.4 快取更新 10
2.2 請求路由 11
2.2.1 請求路由機制 13
2.2.2 請求路由演算法 15
2.2.3 平衡負載裝置 16
2.2.4 重導向裝置 17
第三章 CDN系統設計 19
3.1 系統架構與設計 19
3.2 重新導向演算法 20
3.2.1 Surrogate bandwidth-based之頻寬監控機制 21
3.2.2 RTT-based之頻寬監控機制 23
3.2.3 服務交錯 23
第四章 模擬環境與效能評估 30
4.1 模擬參數與網路架構 30
4.1.1 模擬參數：Surrogate bandwidth-based之頻寬監控 30
4.1.2 模擬參數：RTT-based之頻寬監控 31
4.2 模擬結果與效能評估 32
4.2.1 模擬結果與效能評估：Surrogate bandwidth-based 33
4.2.2 模擬結果與效能評估：RTT-based 35
第五章 結論與未來展望 42
5.1 結論 42
5.2 未來展望 42
參考文獻 44

[1]J. Ni, and D. H. K. Tsang, “Large-Scale Cooperative Caching and Application-level Multicast in Multimedia Content Delivery Networks,” IEEE Communications Magazine, Vol. 43, Issue. 5, pp. 98-105, May 2005.
[2]A. K. Pathan, and R. Buyya, “A Taxonomy and Survey of Content Delivery Networks,” Tech Report, Univ. of Melbourne, 2007.
[3]J. Kangasharju, J. Roberts, K. W. Ross, "Object Replication Strategies in Content Distribution Networks," Computer Communications, vol. 25, no. 3, pp. 367-383, March 2002.
[4]D. Katsaros, G. Pallis, K. Stamos, A. Vakali, A. Sidiropoulos, and Y. Manolopoulos, "CDNs Content Outsourcing via Generalized Communities," IEEE Trans. Knowledge and Data Engineering ,Vol. 21, Issue. 1, pp. 137-151, January 2009.
[5]K. Stamos, G. Pallies, and A. Vakali, Caching Techniques on CDN Simulated Frameworks, Berlin/Heidelberg:Springer-Verlag, 2008, ch5.
[6]A. Vakali, and G. Pallis, “Content Delivery Networks: Status and Trends,” IEEE Internet Computing, Vol.7, Issue. 6, Nov.-Dec. 2003, pp. 68-74.
[7]T. Wauters, J. Coppens, B. Dhoedt, P. Demeester, "Load balancing through efficient distributed content placement," Next Generation Internet Networks, April 2005, pp. 99-105.
[8]N. Bartolini, E. Casalicchio, and S. Tucci, "A Walk through Content Delivery Networks," Lecture Notes in Computer Science, Volume 2965, pp. 1-25, January 2004.
[9]M. Day, B. Cain, G. Tomlinson, P. Rzewski, "A Model for Content Internetworking (CDI)," Internet Engineering Task Force, RFC 3466, February 2003.
[10]A. Barbir, B. Cain, R. Nair, O. Spatscheck, "Known Content Network (CN) Request-Routing Mechanisms," Internet Engineering Task Force, RFC 3568, July 2003.
[11]P. Rzewski, M. Day, D. Gilletti, "Content Internetworking (CDI) Scenarios," Internet Engineering Task Force, RFC 3570, July 2003.
[12]林裕強著，「在內容傳遞網路中的內容配置機制的設計」，碩士論文，國立臺北科技大學資訊工程系碩士班，台北，2011。
[13]劉之揚著，「使用熱門廣播法及支援點對點傳輸之影音內容傳遞網路」，碩士論文，國立中央大學資訊工程系碩士班，2003。
[14]謝民一著，「複製可追查型CDN-P2P遞送網路下之近端伺服器選擇」，碩士論文，國立中央大學資訊工程系碩士班，2004。
[15]Guillaume Pierre and Maarten van Steen, "Globule: A Collaborative Content Delivery Network," IEEE Communications Magazine, Vol. 44, Issue. 8, pp. 127-133, August 2006.
[16]F.L. Presti, C. Petrioli and C. Vicari, "Distributed Dynamic Replica Placement and Request Redirection in Content Delivery Networks", 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, pp. 366-373, Oct. 2007.
[17]Cranor C., et al., "Enhanced Streaming Services in a Content Distribution Network," IEEE Internet Computing, Vol. 5, Issue. 4, pp. 66–75, Jul./Aug. 2001.
[18]OMNeT++, URL :http://www.omnetpp.org/.