由於近年來WWW快速成長為網際網路上最廣泛且使用者最眾的資訊系統，網站建置者大幅投資提升伺服器效能，加上Server Farm架構的出現，使得Web伺服器能力突破單機極限後更是突飛猛進，造成傳統網頁效能測試器的測試成本必須大幅提高才能追上伺服器端的效能。再者，由於動態網頁技術與資料庫關連被廣泛應用，使得帶有狀態與身份的HTTP request（如session）等大量增加，相應而生的各種新式傳輸方法也不在傳統Web Benchmark支援之列。對於測量的結果，傳統Web Benchmark只報告概觀性的最大值與平均值的量化表示也無法適當的表達現今含帶大量動態網頁運作的伺服器效能。以上種種問題，已經開始凸顯了傳統Web Benchmark的不足。
在本論文中，我們針對上述問題，設計與實作一個可組態且具成本效益的網頁效能測試器，我們使用「先定劇本（Workload）」的概念，事先製作好詳細的HTTP request table，可以指定發送時間、內容與使用的HTTP功能，並利用可組態與置換的開放模組設計，讓Workload產生器、Load產生器、報表產生器三大群模組互相獨立，甚至可以分離成三個獨立程式，使得本Benchmark易於適應新技術而不需更動核心部分。我們也採用虛擬使用者概念來模擬一個實體的使用者行為，透過Cookie自動應答與request依使用者分配於相同處理程序使的HTTP狀態與身份資訊得以延續。為了提升成本效益，我們會針對每台Load產生器的系統效能作自我測量並量化成數值，並且依照各台測試機回報之數值來適當分配Workload，使各台測試機可盡其所能，不至於被弱者拖累全體或因為過度負載而導致量測誤差太大。而測量後的結果展示，我們的設計可以依據時間軸演進來描述伺服器效能與負荷的變化，可與request類別跟URL相對照，讓管理者更易於明瞭造成負荷瓶頸的原因與時間點。綜合來看，我們的Web benchmark在功能性方面不輸給傳統Web benchmark，且更有勝過之處。

As WWW service grow up rapidly and becomes the most popular information system on the internet, web site owner invest heavily to improve the performance of web server. And, because of the 'Server Farm' architecture comes real, web server performance breaks the limit of single server, which directly cause server performance improvement. All these factors raise the cost of website performance measurement program to catch up server performance. Besides, because of dynamic webpage and database linkage are applied widely in production environment. HTTP requests (for example, sessions) with state and identifications increase rapidly. Traditional Web Benchmarking methods are out of date and not supporting these new transaction models. Also, traditional benchmarking methodologies only provide the max/average values for overview, which can not properly describe the performance of the websites with massive dynamic webpages. All these problems shows the traditional benchmarking methodologies are not sufficient for today's technologies.
In this paper, we have designed and implemented a configurable and cost-effective website performance measurement program against these problems. We introduced the concept of 'workload', and pre-design detailed HTTP request table, where time, contains, and HTTP commands to use can be assigned. Also we use configurable and replaceable design of open modules and divide the major modules into workload generator, load generator and report generator. These modules can be even separated into three independent programs, which makes this benchmark program become more flexible and adaptive to fit new technologies without adjusting the kernel. We also introduced the concept of 'Virtual User' to describe a real user behavior. We could keep the HTTP state and identification by automatic reply via cookie and assigning user identification in the same process. To increase the efficiency, each load generator can do self-diagnostics and quantify the measurements, and properly reassign the workloads by the value returned by the system. These makes every load generator do everything it could, and not been halted by the low-speed machine. And we can also prevent the mis-measurement by overloading. From the result of the experience, our design can describe the web server performance and load changes by time. We can also compare it to request category and URL, to show the root of causes and time basis to administrator. Overall, our web benchmarking methodology shows the strength to traditional web benchmarking methods.

第一章、緒論 8
1.1 研究背景 8
1.2 研究動機與目的 9
1.3 章節導讀 9
第二章、背景分析 11
2.1 關於WWW 11
2.2 目前Web Benchmark面臨的問題 13
2.2.1 測量功能上的不足 13
2.2.2 Web Server效能量化描述的困難 13
2.2.3 低成本、高效能 14
第三章、設計與架構 15
3.1 基本設計概念 15
3.1.1 先定劇本 15
3.1.2 虛擬使用者 16
3.1.3 依照Load Generator能力分配Workload 16
3.1.4 Web效能的量化表示 16
3.2 Client/Server架構 17
3.3 可組態之模組設計 18
3.3.1 Controller程式之模組設計 19
3.3.2 Load Generator程式之模組設計 20
3.3.3 可組態且可置換之模組設計 21
3.4 Workload Generator劇本產生器 22
3.5 Session support 22
3.6 Workload分配 23
第四章、網頁效能測試器之實作 24
4.1 組態設定檔 24
4.2 時間之同步 25
4.3 Workload劇本（request table）結構 25
4.4 Work Log結構 26
4.5 Workload分配機制 27
4.5.1 Load Generator機器系統效能測量 27
4.5.2 Workload分配方法 28
4.5.3 避免Load Generator過度負載的檢查 29
4.6 Web log replay模組 30
第五章、測試結果與分析 32
5.1 測試目的 32
5.2 測試環境 32
5.3 測試結果 32
5.3.1 不同發送頻率下的結果 33
5.3.2 特定種類request的結果 34
5.3.3 CPU負荷的觀察 34
第六章、相關研究與實作之比較 36
6.1 Zeus Benchmark 36
6.2 WebStone Benchmark 36
6.3 WebBench Benchmark 37
6.4 SPECWeb99 Benchmark 37
6.5 綜合評論 38
第七章、結論與未來方向 39
7.1 結論 39
7.2 未來方向 40
參考文獻 42

[1] T. Berners-Lee, R. Cailliau, A. Luotonen, H. Nielsen, A. Secret. “The World-Wide Web” Communications of the ACM Aug. 1994.
[2] Henning Schulzrinne. “World Wide Web : Whence, Whiter, What next ?” IEEE Network, March 1996.
[3] HTTP - Hypertext Transfer Protocol Overview, http://www.w3.org/Protocols/
[4] “Universal Resource Identifiers in WWW : A Unifying Syntax for the Expression of Names and Addresses of Objects on the Network as used in the World-Wide Web”, RFC-1630, CERN, June 1994.
[5] T. Berners-Lee and D. Connolly, “HyperText Markup Language Specification – 2.0”, November 1995.
[6] Dave Raggett, “HyperText Markup Language Specification Version 3.0”, IETF Internet Draft, March 1995, RFC 1945.
[7] T. Bererns-Lee, R. Fielding, H. Frystyk, "Hypertext Transfer Protocol -HTTP/1.0", RFC-1945, MIT/LCS, UC Irwine, May 1996
[8] R. Fielding, J. Gettys, J.C. Mogul, H. Frystyk, T. Berners-Lee, "Hypertext Transfer Protocol - HTTP/1.1", RFC-2068, UC Irvine, Ditigal Equipment Corporation, MIT.
[9] The Common Gateway Interface, http://hoohoo.ncsa.uiuc.edu/cgi/
[10] D. Kristol, L. Montulli, “HTTP State Management Mechanism”, RFC-2109, February 1997.
[11] SPECweb99 Benchmark, http://www.spec.org/osg/web99/
[12] WebStone Benchmark, http://www.mindcraft.com/webstone/
[13] WebBench, http://www.webbench.com/
[14] Zeus benchmark in Apache package, http://www.apache.org/
[15] J. Hu, S. Mungee, and D.s C. Schmidt. Techniques for Developing and Measuring High-Performance Web Server over High Speed Networks. In Proceedings of the IEEE Infocom’98 Conference, San Francisco, CA, April 1998.
[16] S. Manley, M. Seltzer, and M. Courage. A Self-Scaling and Self-Configuring Benchmark for Web Servers. In Proceedings of the ACM SIGMETRICS’98 Conference, June 1998.
[17] P. Barford and M.E. Crovella, “Generating representative Web workloads for network and server performance Evaluation”, in Proceedings of Performance’98/ACM SIGMETRICS’98, pp.151-160, Madison WI.
[18] M. Arlitt, T. Jin. “A Workload Characterization Study of the 1998 World Cup Web Site”. IEEE Network, May/June 2000.
[19] Crovella, M., and A. Bestavros, "Self-Similarity in World Wide Web Traffic: Evidence and Possible Causes," Proc. ACM Sigmetrics '96, Philadelphia, May 1996.
[20] 羅孟彥，具高可及性與可擴充性Web伺服器之設計與實作，國立中山大學資訊工程研究所碩士論文
[21] RFC 2069. Proposed Standard. An Extension to HTTP: Digest Access Authentication, Jeffery L. Hostetler, John Franks, Phillip Hallam-Baker, Paul Leach, Ari Luotonen, Eric W. Sink, Lawrence C. Stewart, January1997.
[22] http://www.rdec.gov.tw/ipcs/http.htm