Measurement tools and modeling techniques for evaluating web server performance
The past few years have seen a rapid growth in the popularity of the Internet and the World Wide Web. Many companies are deploying Web servers and seeing their usage rates climb rapidly over time. Our research focused on analyzing and evaluating the performance of Internet and intranet Web servers with a goal of creating a Layered Queueing Model to allow capacity planning and performance prediction of next generation server designs. To achieve this we built a tool framework that enables us to collect and analyze empirical data necessary to accomplish our goals.
This paper describes custom instrumentation we developed and deployed to collect workload metrics and model parameters from several large-scale, commercial Internet and intranet Web servers over a time interval of many months. We describe an object-oriented tool framework that significantly improves the productivity of analyzing the nearly 100 GBs of collected measurements. Finally, we describe the layered queueing model we developed to estimate client response time at a Web server. The model predicts the impact on server and client response times as a function of network topology and Web server pool size.
KeywordsNetwork Delay Capacity Planning Common Object Request Broker Architecture Queueing Network Model Server Response Time
Unable to display preview. Download preview PDF.
- World-Wide Web: The Information Universe. T. Berners-Lee, R. Cailliau, J. Groff, B. Pollermann. Electronic Networking: Research, Applications and Policy, v2nl. Spring 1992.Google Scholar
- Web Server Workload Characterization. J. Dilley. HPL TR 96-160, December 1996.Google Scholar
- Measurement Tools and Modeling Techniques for Evaluating Web Server Performance (unabridged). J. Dilley, R. Friedrich, T. Jin, J. Rolia HPL TR 96-161, December 1996.Google Scholar
- Web Server Workload Characterization: The Search for Invariants. M. Arlitt, C. Williamson, University of Saskatchewan. In ACM SIGMETRICS, Philadelphia, May 1996.Google Scholar
- A Synthetic Workload Model for Internet Mosaic Traffic. M. Arlitt, C. Williamson, Discus Working Paper 95-8, University of Saskatchewan.Google Scholar
- User Access Patterns to NCSA's World Wide Web Server. T. Kwan, R. McGrath, D. Reed, University of Illinois. IEEE Computer, Vol 28, No. 11, November 1995.Google Scholar
- A Toolset for Performance Engineering and Software Design of Client-Server Systems, G. Franks, A. Hubbard, S. Majumdar, D. Petriu, J. Rolia, C.M. Woodside, Special Issue of the Performance Evaluation Journal, Volume 24, Number 1–2, Pages 117–135.Google Scholar
- Quantitative System Performance: Computer System Analysis Using Queueing Network Models. E.D. Lazowska, J. Zahorjan, G.S. Graham, K.C. Sevcik, Prentice-Hall, 1984.Google Scholar
- The Standard Template Library. A. Stepanov, M. Lee, Hewlett Packard. ANSI Document No. X3J16/94-0030 WG21/N0417. March 1994.Google Scholar
- STL<ToolKit> Users Guide. ObjectSpace, Inc. C++ Component Series. Version 1.1, May 1995.Google Scholar
- A Queueing Network Analysis of Computer Communication Networks with Window Flow Control. M. Reiser. IEEE Transactions On Communications, August 1979, pp. 1201–1209.Google Scholar
- Common Object Request Broker Architecture and Specification (CORBA). Object Management Group Document Number 91.12.1, Revision 1.1.Google Scholar
- OSF DCE. H. Lockhart. McGraw Hill, 1994. ISBN 0-07-911481-4.Google Scholar