This documentation specifies the public API for controlling offline profiling , accessing the JProfiler MBean , writing embedded probes and injected probes as well as the profiling platform that is part of JProfiler. See the overview pages of the respective root packages that are linked above for more information. Contains the controller class for programatically controlling the recording of profiling data. Contains classes that represent parameters for requesting profiling data from a connection.
|Published (Last):||10 January 2011|
|PDF File Size:||2.23 Mb|
|ePub File Size:||1.60 Mb|
|Price:||Free* [*Free Regsitration Required]|
Please see the change log for a detailed list of changes. What's New in JProfiler 4. With the profiling platform you can write your own custom profiler for special situations. You can profile remote JVMs or load snapshots and analyze them programmatically with the profiling platform.
In addition, you can also run a custom profiler in the same process - applications of this scenario include advanced monitoring facilities. The API is written for Java 1. The interceptor API allows you to inject code at entry and exit of arbitrary methods, giving you access to method arguments and return values.
With the measurement you can create "payload" objects and add them to the call tree. Each interceptor creates its own hot spot type. The screenshot below shows payload data in the call tree from that interceptor:.
The hot spot type contributed by an interceptor allows you to analyze payload from all call stacks grouped by the payload description:. EJB 3 support. The class tracker enables you to observe how instance counts for selected classes or packages develop over time.
This expands on the existing differencing columns in other memory views. It is a highly useful tool for observing the development of classes whose instance counts are critically high or continuously rising due to a possible memory leak. The selection of the tracked classes as well as the liveness and object type options are persistent for each session. The integration has been available on the plugin server since the EAP and is now bundled with this JProfiler release.
Netbeans 5. The Netbeans integration now gives access to the snapshot comparisons feature and supports profiling non-bundled Sun Java System Application servers.
Several bugs in the integration were fixed. For early adopters, JProfiler tries to support milestones in a timely fashion. This release supports eclipse 3.
An action to highlight the shortest path to the main node in the reference graph of the heap walker. This action highlights the best path in red which allows you to quickly find your way from any node to the main object that the reference graph is currently displaying.
An action to select java. Class objects in the Classes view of the heap walker. This action is especially useful if you suspect a classloader-related memory leak. With the resulting class objects, you can directly go to the reference graph and search for a path to a garbage collector root to check why a class has not been garbage collected. Optionally calculate snapshot CPU comparisons with average call times. This new option uses total times divided by invocation counts and is available for the CPU call tree comparison as well as the CPU hot spot comparison.
Layouters in the Call Graph view. Previously only the hierarchic layouter was available in the call graph. The new choice of layouters allows you to present the call graph in other, possible more informative visualizations. Previous: JProfiler Version 4. Need server monitoring? Have a look at JProfiler's sibling product perfino. Latest Versions.
JProfiler Blog Posts
JProfiler API documentation
It is developed by ej-technologies and currently in version JProfiler provides many IDE integration and application server consolidation. For example, the number of objects, the size, the method execution stack created by the object, the hot spot created by the object. Heap walker: The memory collected for a certain period of time is static analysis of information? Threads: This module is mainly used to monitor the running state of the management thread.
Please see the change log for a detailed list of changes. What's New in JProfiler 4. With the profiling platform you can write your own custom profiler for special situations. You can profile remote JVMs or load snapshots and analyze them programmatically with the profiling platform. In addition, you can also run a custom profiler in the same process - applications of this scenario include advanced monitoring facilities. The API is written for Java 1.