Performance Analysis and Performance Modeling of Web-Applications

Abstract

We study the performance of coupled web servers and database servers of certain web-applications. We focus on the performance bottleneck of one of the most pressing applications and their average response time. Therefore we present the software architecture of this web-application, called "Study-Portal" of the University of Mannheim, and we identify typical workload scenarios. We discuss our performance measurements and document the software and hardware infrastructure used during the measurements. Our goal is the identification of the most influential parameters which determine the timing behavior of the combined web-application. For the modeling we will use standard stochastic methods, even if it will mean over simplification. Our results then provide a simple estimate for the considered web-application: a doubling of application servers leads to half of the response time in case of sufficiently many clients. The developed model fits well with the observations in the operating environment of an IT-center.