隨著網路的發達，使用網路的人口越來越多，網路也越來越充滿商機，很多廠商都在網路上提供各式各樣的服務供民眾使用，而為了方便使用，大部分的服務都是使用網頁的方式呈現，網頁伺服器的效能就是能否提供良好服務的關鍵之ㄧ，以使用者的角度來看，網頁伺服器效能取決於伺服器的回應時間，所以本論文實際建立一個被動式量測網頁伺服器回應時間的系統。

本論文並建立一個實驗環境來探討當伺服器端遇到瓶頸時會造成什麼現象，伺服器端可能存在的瓶頸包括處理器、網路與硬碟存取，我們設計了幾個實驗來造成伺服器面臨上述所說的各種瓶頸，觀察在這些狀況下會有什麼現象產生，從實驗結果中發現在各種瓶頸下，各種要求類型與回應時間會呈現不同的現象，由此可知從中間端來得出伺服器狀態是可行的。

Because of the development of the Internet, there are more and more users. Various kinds of services are provided by companies. They are usually web based for easy use. The
performance of web servers is a key factor of the quality of services. In a user’s view, the response time is a important metric for the performance of a web server, so we implement a passive performance evaluation tool for web servers in this paper.

In this paper we discussed phenomenons when bottlenecks at a web server occur. We make experiments to show how bottlenecks in the CPU, network, and in the Disk can affect the response time. The result shows that response time is affected in different way according to the type of the request when the bottleneck is CPU, network or disk.

第一章緒論
1.1 研究動機與目標
1.2 章節導讀
第二章相關研究
2.1 主動式量測方法
2.2 被動式量測方法
第三章系統架構與實作
3.1 系統位置
3.2 系統架構
第四章實驗方法與分析
4.1 數值的準確性
4.2 CPU 瓶頸實驗
4.2.1 實驗方法
4.2.2 實驗結果
4.3 網路瓶頸實驗
4.3.1 實驗方法
4.3.2 實驗結果
4.4 Disk I/O 實驗
4.4.1 實驗方法
4.4.2 實驗結果
4.5 從實驗中得到的啟發
第五章結論與未來工作

[1] Netcraft: Web Server Survey Archives. http://news.netcraft.com/archives/web_server_survey.html.
[2] Fluke Networks: SuperAgent. http://www.flukenetworks.com/us/LAN/Monitoring+Analysis+Diagramming/SuperAgent/Overview.htm.
[3] SPECweb2005. http://www.spec.org/web2005/.
[4] Technorati. http://www.technorati.com/.
[5] The Apache HTTP Server Project. http://httpd.apache.org/.
[6] TPC-W. http://www.tpc.org/tpcw/.
[7] WebStone. http://www.mindcraft.com/webstone/.
[8] Banga, G., and Druschel, P. Measuring the capacity of a web server. In USENIX Symposium on Internet Technologies and Systems (1997), 61–71.
[9] Dilley, J., Friedrich, R., Jin, T., and Rolia, J. Web server performance measurement and modeling techniques. Perform. Eval. 33, 1 (1998), 5–26.
[10] Fielding, R., Gettys, J., Mogul, J., Frystyk, H., Masinter, L., Leach, P., and Berners-Lee, T. Hypertext transfer protocol – http/1.1. RFC2616 (1999).
[11] Fu, Y., Vahdat, A., Cherkasova, L., and Tang, W. Ete: Passive end-to-end internet service performance monitoring. In Proceedings of the General Track:2002 USENIX Annual Technical Conference (Berkeley, CA, USA, 2002), USENIX
Association, pp. 115–130.
[12] Gschwind, T., Eshghi, K., Garg, P. K., and Wurster, K. Webmon: A performance profiler for web transactions. In WECWIS ’02: Proceedings of the Fourth IEEE International Workshop on Advanced Issues of E-Commerce and Web-Based Information Systems (WECWIS’02) (Washington, DC, USA, 2002), IEEE Computer Society, p. 171.
[13] Jacobson, V., Leres, C., and McCanne, S. Tcpdump public repository. http://www.tcpdump.org/.
[14] Mosberger, D., and Jin, T. httperf – a tool for measuring web server performance.SIGMETRICS Perform. Eval. Rev. 26, 3 (1998), 31–37.
[15] Olshefski, D. P., Nieh, J., and Nahum, E. ksniffer: Determining the remote client perceived response time from live packet streams. In Proceedings of the Sixth Symposium on Operating Systems Design and Implementation (2004).
[16] Slothouber, L. P. A model of web server performance. In Proceedings of the 5th International World Wide Web Conference (1996).
[17] Urgaonkar, B., Pacifici, G., Shenoy, P., Spreitzer, M., and Tantawi, A. An analytical model for multi-tier internet services and its applications. In SIGMETRICS ’05: Proceedings of the 2005 ACM SIGMETRICS international conference on Measurement and modeling of computer systems (New York, NY, USA, 2005), ACM Press, pp. 291–302.