Insight into Java application performance bottlenecks: Visual performance Analyzer

Using VPA to quickly locate Java application performance bottlenecks

Introduction: Large Java applications call a large number of classes and methods, how to find the application performance bottlenecks in the thousands of lines of code? In this article, the authors describe how to select Profiling Tools for different performance issues, sample performance issues, and use Visual performance Analyzer to analyze performance sampling results and quickly locate performance bottlenecks.

Introduction

Classes are the foundation of Java. Large-scale Java applications are made up of thousands of classes. When performance problems arise, how can you quickly locate performance bottlenecks in this large heap of classes? To make things worse, some of the classes were written by a colleague in the last century, some of which were provided by third parties, and no one understood what the performance impact of these classes would be for the entire application.

About VPA

VPA (visual performance Analyzer) is a set of visualization profiling Tools based on Eclipse development.

VPA provides six pieces of weapon for performance analysis: Profile Analyzer, Code Analyzer, Pipeline Analyzer, Counter Analyzer, Trace Analyzer, call Tree Analyzer. Each tool can be used to analyze a specific category of performance data.

If you want to learn more about VPA, you can refer to the VPA User's Guide.

The two tools in VPA: Profile Analyzer and CallTree Analyzer provide strong support for performance analysis of Java applications. These two tools provide multiple views that help users analyze performance data from different angles. With the help of these two tools, users can quickly find the worst-performing methods or lines of code from the thousands of lines of code.

The process of using VPA to analyze performance problems can be divided into three steps:

For different performance problems, select the appropriate performance analysis tools;

collecting performance data;

Use VPA to analyze performance data;

This article will take applet bookstore as an example to describe how to use these two tools to quickly locate performance issues.