Why Is Performance Testing and Tuning Important?

When designing an application that will be accessed over the Internet, development teams must keep in mind that hundreds of thousands of users could concurrently access the application. This concurrent user load could place tremendous stress upon a system, cause unexpected delays, and result in a poor user experience. However, if the application has been adequately tested and tuned for optimal performance under stress, managers and software developers will be confident that their code will perform at optimal levels and have the necessary data to plan site capacity needs. The following real world example demonstrates what can happen when adequate performance testing has not occurred.

Real World Example Site Traffic Spikes

On September 11, 2001, news events of attacks on the World Trade Center and the Pentagon caused site traffic at a Microsoft Web news site to increase by threefold compared to pre-September 11 levels. In the days following the attack, the increased level of traffic persisted as the public demand for news persisted. This caused the page response times on this particular site to degrade to unacceptable levels, and the Web servers required rebooting every couple of hours as available memory became critically low. It was subsequently determined that there was an existing memory leak associated with this site, which was exacerbated by the sudden increase in traffic. The methodology presented in this book is designed to proactively test and subsequently prevent unacceptable Web site performance degradations, even in the face of unexpectedly large traffic spikes, which can be caused by extraordinary events such as those on September 11, 2001.

This example should lead to the question, How does one approach the task of accurately performance testing and tuning a Web application? Unfortunately, there isn t one correct answer to this question. Methodologies aimed at performance testing and tuning Web applications are constantly evolving, and new tools for obtaining a more accurate result set are quickly becoming available. In the case of a simple Web site, a set of static stress scripts might do the job; whereas, in the case of a complex e commerce site, performance modeling tools and dynamic stress scripts would be required to better predict how the application would react under a particular user load. Because acceptable Web site performance is application dependent and Web site architecture can vary widely from site to site, Web site performance analysis can be viewed as more an art than a science. The goal of effective performance testing, however, is to ensure that the site will perform within acceptable performance levels, whatever those levels need to be.