Chapter 13: Performance Management Tooling

In previous chapters, I discussed ways to implement WebSphere applications and environments with performance and availability in mind. In this chapter, you'll explore some methods and approaches for monitoring and analyzing system and application data so you can understand the performance and load characteristics of your WebSphere servers and applications.

Overview of Performance Tooling

Performance tooling is the fairly broad name given to the process of selecting, implementing, and using performance analysis tools to monitor and probe your applications and servers for information on their load and performance status. In this chapter, you'll look at how to measure some key performance aspects of your systems and correlate those findings with the perceived performance of your WebSphere applications.

That is, you'll look at several key areas of a server that can impact the performance of a WebSphere application. I'll discuss this from both a proactive and reactive point of view. In other words, if your WebSphere applications appear to be running slowly, what steps will you take to understand what's going within your WebSphere application?

Many of you may already know about some or all of these tools if you come from a system administrative background. However, what I've tried to do in this chapter is match the tools to the problem area they can fix and provide examples of what the tools' output really means with regard to your WebSphere applications.

Specifically, you'll see a range of commands and utilities that are useful in determining system and application load characteristics. The first part of the chapter discusses the tools that are best suited to diagnosing and understanding the system state (load, utilization, and so on). At the end of these sections, you'll explore which tools are best for diagnosing potential performance issues and learn how to read the results of the diagnosis. I'll use tables to summarize possible causes to your performance problems. Each scenario will include one or more possible reasons for the performance problem and a guide to hunting down and diagnosing the performance issue.

Essentially, performance tooling is a part of the performance management methodology. It's when you use tools and applications to understand what's going on within your WebSphere application. Essentially, without performance tooling, you'll never know what may be causing a system to operate inefficiently. Many of the tools used for performance tooling are available as part of each operating system or are readily available on the Internet.

In this chapter, you'll look at the following:

• Monitoring system utilization ”for example, Central Processing Unit (CPU), memory, and swap space

• Monitoring disk Input/Output (I/O) utilization

• Monitoring network performance and utilization

• Monitoring hot spots within WebSphere applications

At the end of this chapter, you should be able to run tools and understand, either reactively or proactively, the state of your applications via the state of your servers.