論文摘要
主要研究users與servers之間的互動行為，我們將稱它是”Session-Based” 的模式，再對session-based加入我們想要的policies，通稱為 ”Session-Based Management”。對於網路發展多元化快速的成長，以及寬頻網路世界的來臨，網路所能提供的service種類越來越豐富，使用者對於Internet Service的要求品質也相對提昇，為了能夠提供好的Internet Service（例如：e-commerce site等等 …）品質，了解使用者的需求所在，以及了解使用者的使用習慣，我們認為這是一個值得研究及分析的方向。

首先我們運用以及研究目前使用於網際網路間，在Client-Server的架構中，用來記錄以及認識用戶端資訊最廣泛的技術 ”cookie”。針對cookie的設計，使得當每一用戶端連上網站的時候，可以輕易的識別出該用戶端的資訊，並針對該用戶端的特性，給予適當的回應或提供應有的服務品質。另外，對於目前現行網站中（例如：eBay , 買賣王 , kimo , openfind）content使用cookie的技術，也是我們想要知道的方向，因此我們進行分析的方向包括：將Web Service的性質分類，例如 – Shopping cart、Search engine、競標網站、會員制登錄網站等等…，我們必須了解的是這些Web Service對於cookie技術的應用，以及對於Client的impact有多大。接著，我們先實作一個EC site，包括Database的一個three tier架構，再寫程式去對流經這個site的所有封包做分析，從底層將封包取出，經過我們事先寫好的程式將整個封包架構用我們的方式表達出來，再抓取出我們所需要的資料（例如：URL , cookie … ），再針對分析封包的結果來制定我們Session-Based的policies。

我們訂定我們環境在Server Cluster的架構，現今的網站架構中，僅剩一些小型的企業網站仍然使用傳統單一的Client-Server架構，或是將它們的網站拖管，越來越多的ISP（Internet Service Provider）或是ICP（Internet Content Provider）則是將他們的服務架構在Cluster Servers的架構中（例如：Yahoo , eBay , Netscape 等等 …）或是稱為Hosting Service的架構。因此，在這些異質架構的server組合之下，我們前端的distributor，要面對的是不同web servers所製造出來的網頁內容給不同的使用者，透過追蹤使用者的資訊，就能夠提供servers對使用者的更加了解，進而符合使用者的需求，達到提供使用者良好服務品質的目的，提供使用者 session-based differentiated service，也提供不同使用者在session-based的priority，我們則是將differentiated service 做在我們的session-based之上，並實際運用在我們的Server Cluster架構中。

Abstract
We propose to research the interaction between users and web servers in Internet , and we call that “Session - Based model“ . Then we add some policies to the session-based model , we define that is “Session - Based Management“ . For the explosive growth of Internet service（eg.,e-commerce…）, we consider about what users want and tracing users’ behaviors , those are what we want to research and analysis .
We using “cookie“ , what is the technique to use on the Client-Server model of Internet . This make server directly and easily know some information about users . So server supply quality of service to user what they have identified . Then we trace technologies of cookie which are used in some popular web sites（eg., eBay , ubid , kimo , openfind）, and analyze the impact of users about those technologies .
For example , we classify the Internet service to – shopping cart、search engine… Finally , we construct our e-commerce web site to parsing every packets through our site , getting the information what we want from those packets , and then we define our some policies into our Session-Based model .
The infrastructure of our implementation environment is in Server Cluster Architecture , which is the most popular one this time . More and more ISPs（Internet Service Provider）or ICPs（Internet Content Provider）construct their web sites in this kind of structure . In our cluster system , our distributor will analyze all kinds of packets from all heterogeneous servers . Using our technique will make distributor to learn how to know the user information and identify the users , so distributor will know how to supply users’ session-based priority、differentiated service and so on . We actually implement our mechanisms in our server cluster system .

第一章 簡介 3
1.1 簡介 3
1.2 動機 5
第二章 相關研究探討 6
第三章 進行步驟集方法 7
3.1 研究環境 7
3.2 進行步驟 7
3.3 進行方法 8
3.3.1. 架設實驗性Web site 8
3.3.1.1.利用傳統的Cookie技術
3.3.1.2.利用PHP session function
3.3.1.3.利用database暫存使用者狀態
3.3.2. Cookie之相關研究 10
3.3.3. 現行網站之相關研究 13
3.3.3. Packets log之相關研究 16
第四章 設計與實作 20
4.1 辨識session 20
4.2 辨識session requests 22
4.3 做到differentiated service 24
4.4 量測結果 26
第五章 結論與未來研究方向 31
參考文獻 32

[1] M. Y. Luo. And C. S. Yang. “System Support for Scalable, Reliable and High Manageable Web Hosting Service”[2]C. S. Yang. and M. Y. Luo. “Realizing Fault Resilience in Web-Server Cluster “
[3]Yahoo,http://www.yahoo.com
[4]Altavista,http://www.altavista.com
[5]eBay,http://www.ebay.com
[6]Netscape, http://www.netscape.com
[7]Cherkasova, L. and Phaal, P. “Session Based Admission Control : a Mechanism for Improving Performance of Commercial Web Sites”
[8]DoubleClick,http://www.doubleclick.com
[9]Thomas, B. “Recipe for e-commerce”
[10] [RFC 2109]“HTTP State Management Mechanism”
[11]Stathes P. Hadjiefthymiades and Drakoulis I. Martako. “Improving the performance of CGI compliant database gateways”. In Proc. Of the 6th Int’l World Wide Web Conference (WWW6). April 1997.
[12] [RFC 2068] “Hypertext Transfer Protocol -- HTTP/1.1”
[13] Ghaleb Abdulla, Edward A. Fox, and Marc Abrams. Shared User Behavior on the World Wide Web. In WebNet World Conference of the WWW, Internet, and Intranet, Orlando, Florida, USA , November 1997. http://www.cs.vt.edu/~chitra/docs/97webnet/.
[14] M. Arlitt. Characterizing Web User Sessions. In Proceedings of the Performance and Architecture of Web Servers Workshop, Santa Clara, California, June 2000.
[15] G. Bilchev, I. Marshall, S. Olafsson, and C. Roadknight. Modelling Http Traffic Generated by Community of Users. In High Performance Computing and Networking Conference, Amsterdam, Netherlands , April 1999.
[16]Netscape Communications Corporation Client Side State-HTTP Cookies http://www.netscape.com/newserf/std/cookiespec.html
[17]D. A. Menasce et al."A Methodology for Workload Characterization of E-Commerce Sites," Proc. 1999 ACM Conf. Electronic Commerce, New York, 1999
[18] Wang, Z.; Basu, A.Resource allocation for elastic traffic: architecture and mechanisms, Network Operations and Management Symposium, 2000. NOMS 2000. 2000 IEEE/IFIP , 2000
[19] Ingham, D.B.; Shrivastava, S.K.; Panzieri, F. Constructing dependable Web services, IEEE Internet Computing , Volume: 4 Issue: 1 , Jan.-Feb. 2000
[20] Ho, S.; Loucks, W.M.; Singh, A. Monitoring the performance of a Web service, Electrical and Computer Engineering, 1998. IEEE Canadian Conference on , Volume: 1, 1998
[21]Cherkasova Ludmila and Peter Phaal. Session based admission control: A mechanism for improving the performance of an overloaded web server, HP Labs Technical Reports, External HPL-98-119, 980612, Hewlett Packard, June 1998. 22
[22]Ludmila Cherkasova, Ludmila Cherkasova and
Peter Phaal. Session Based Admission Control: a Mechanism for Web QoS, - Hewlett-Packard Laboratories 1501 Page Mill Road, Palo Alto, CA 94303
[23]D. Krishnamurthy and J. Polia. Predicting the Performa-
nce of an E-Commerce Server : Those Mean Percentiles. in
Proc. First Workshop on Internet Server Performance. ACM
SIGMETRICS 98. June 1998