Semester

Fall

Date of Graduation

2005

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Katerina Goseva-Popstojanova.

Abstract

The World Wide Web has been a huge success, bringing the Internet to widespread popularity. For Web based systems to deal effectively with increasing number of Web clients, it is very important to understand the basic fundamentals of Web workload and error characteristics. In this thesis we focus on detailed empirical analysis of Web server error characteristics and reliability based on the data extracted from eleven different web servers. First, we address the data collection process and describe the methods for extraction of workload and error data from Web logs. Then, we analyze the Web error characteristics which include unique errors, frequency of occurrence of unique errors and top files causing errors. Furthermore, we analyze the relationship between errors among Web workload and estimate request-based and session-based reliability. The discussion presented in this thesis shows the sessions-based reliability is better indicator of user perception of Web quality than request-based reliability. Finally, we analyze and develop heuristic search criteria to identify sessions which indicate unusual server behavior, such as extremely long sessions and sessions with large number of server errors. The results of our study provide valuable measures for tuning and maintaining of Web servers.

Share

COinS