WebBeater
As the speed and popularity of the Internet continues to increase, web servers (computers that serve documents containing information and media to other computers connected to the Internet) must likewise improve in their capability to disperse information quickly and to a wide number of clients (computers requesting information from servers) simultaneously. This applies to all web servers, but most strongly to servers that experience heavy traffic (large numbers of requests from clients).
In order to find ways to improve web server technology, we must first accurately evaluate a server's performance by looking for bottlenecks (components that impede the performance of the rest of the system). This is being accomplished through the development of WebBeater, a set of programs designed to simulate very heavy traffic and effectively beat a web server until it reaches capacity or fails. While WebBeater requests a flood of Internet documents from the server, tools such as hardware trace collection, system monitors, and feedback from WebBeater itself will help pinpoint problem areas in the server's system.
This work is being done primarily by Alen Peacock. Email alen@pel.cs.byu.edu for more information.
Note: WebBeater will be available for public use shortly. For now, here's a screen shot:
WebBeater Master Screen
Status output from the terminal
© 1996, Performance Evaluation Laboratory, Brigham Young University. All rights reserved. Reproduction of all or part of this work is permitted for educational or research use provided that this copyright notice is included in any copy. Send comments to webmaster@pel.cs.byu.edu. This document last modified on Wednesday, 01-Apr-98 14:30:22.