The Java Performance Diagnostic Model (jPDM) is a performance diagnostic model that abstracts the complexity of systems. It helps interpreting the performance counters of the system and thus understanding the root cause of why we experience performance regression.

The challenge with identifying performance regression is that a specific scenario is the result of innumerable influences, many of them external to the application. jPDM and the resulting methodologies helps dealing with that complexity.

In terms of responsiveness, there is an infinite number of things that can go wrong, but they will go wrong in a finite number of ways. Performance regression can therefore be categorized into different manifestations. There will be a few typical forms of issues, emerging in innumerable, varying scenarios, and root causes. In order to identify the different categories, we will make use of the diagnostic model.

jPDM identifies important subsystems of our system, their roles, functions, and attributes. The subsystems interact with each other. The model helps to identify tools to measure levels of activity and interactions between the subsystems. Methodologies and processes that help to study and analyze systems and situations in regard to performance, fall out of this model.