Skip to content


The benefit of wall clock time data in method profiling

As a companion to our overview of the new JFR data adapter in binjr, here’s an illustration of how having good visualization options for the wall clock time component of method profiling events can help gain deeper insights out of a JFR profile, from a recent Real Lifeā„¢ investigation.

In this example, we are faced with an application whose startup time increased dramatically following a migration of some pieces in the software stack, to the point of becoming unbearable.

New Data Adapter: JDK Flight Recorder

It has been a pretty long time since we’ve added support for a new data source in binjr, but coming up in release 3.13 is something quite new indeed: support for recordings produced by JDK Flight Recorder (JFR, for short).

So what, exactly, is JFR? To quote Wikipedia:

JDK Flight Recorder is an event recorder built into the OpenJDK Java virtual machine. It can be thought of as the software equivalent of a Data Flight Recorder (Black Box) in a commercial aircraft. It captures information about the JVM itself, and the application running in the JVM. There is a wide variety of data captured, for example method profiling, allocation profiling and garbage collection related events. The technology was open sourced in 2018.
Analysis and visualization of flight recordings are normally done using JDK Mission Control (JMC).

Every single event recorded by JFR naturally includes a timestamp, making it a pretty straight-forward source of data for binjr, one that can certainly benefit for the freeform navigation and dynamic view composition binjr allows.