透過您的圖書館登入
IP:18.188.44.223
  • 學位論文

P2P封包辨識系統設計與實作

The Design and Implementation of a P2P packets Identification System

指導教授 : 李錫捷
共同指導教授 : 盧以詮(Todger Lu)
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


本論文之P2P封包辨識系統,主要係運用埠號結合封包特徵來進行P2P網路流量的辨識。由於P2P應用程式的出現,造成網路頻寬大量的被P2P流量所佔用,嚴重影響其他網路使用者之網路使用品質,另由於P2P傳遞之檔案通常為非合法授權之檔案,以致有觸法之可能。在資安意識高漲的今日,為能將P2P網路流量進行有效管理,首先需能辨識P2P封包與非P2P封包流量的不同。能有效辨識P2P的網路流量,網路管理人員方可進一步來調整或管制網路頻寬的使用,使得網路品質得以提升。 本研究收集相關P2P封包數據,並根據封包內容進行分類及特徵分析比對,將分析所得之封包特徵以正規表示式進行表示,再依埠號搭配封包特徵,撰寫應用程式來進行自動化封包特徵的比對。為驗證系統之可用性,以實際網路流量進行驗證,擷取企業網際網路流量作為實驗封包數據來源,運用封包特徵辨識方式進行網路流量辨識,以驗證系統是否符合預期結果。並將系統以分散式負載平衡架構進行建構,以驗證系統具延展性,可應付未來網路流量的成長。在驗證系統各項功能後,實驗數據顯示可達良好的運作結果。

並列摘要


The P2P packets Identification system is using the well-known port combine with packet characteristic to detect the P2P network flows. Because of the P2P application appearance, the network bandwidth has been abused by the P2P flows which also make the poor network performance. To manage network flow more effectively, the network administrator first must be able to recognize the difference between P2P flow and non-P2P flow. Then the network administrator may further adjust or control the network bandwidth use to improve network quality. This research collected the P2P packets data, analyzed and classified the data according to the packet characteristic, used the regular expression to present the packet characteristic and then developed an automation packet characteristic recognizing system. The system was built as load balance processing structure and was testified by capture the real enterprise internet network flow.

參考文獻


36. 吳富宇,「具負載平衡機制之虛擬伺服器的設計與實作及其應用」,元智大學資訊管理學系,碩士論文,2004年
27. Robert Kruse, Bruce Leung, and Clovis Tondo, "Data Structures and Program Design in C 2nd", International Edition, pp. 353-364,367-371, 1996.
8. Douglas Comer, "Internetworking with TCP/IP Principles, Protocols, and architectures", Fourth Edition, Volume 1, pp. 98, 197-198, 209-251, 2000.
13. Fulvio Risso, and Loris Degioanni, "An Architecture for High Performance Network Analysis", Proceedings of the 6th IEEE Symposium on Computers and Communications (ISCC 2001), pp. 686-693, July 2001.
19. Mario Baldi, Elena Baralis, and Fulvio Risso, "Data Mining Techniques for Effective Flow-based Analysis of Multi-Gigabit Network Traffic", Computer Engineering Department, October 2004.

延伸閱讀