Forum Stats

  • 3,875,494 Users
  • 2,266,929 Discussions


JMC flight recording code analysis

2768457 Member Posts: 1


I am having trouble understanding my flight recording of an ADF application. I have profiled an ADF app built in 11g R2 running on a stand alone server. The JMC recording returns everything it should.

I need to know how to analyze my ADF methods to address problem areas to increase performance.  The majority of the code being recorded are java methods and not instances of my ADF app. How can I better understand what's going on so I can tweak the code for better performance results?

Thank you,



  • 2753764
    2753764 Member Posts: 1

    I'm working on trying to understand the mission control output a bit better with OP up there.  We have watched the youtube series from Oracle called Java Mission Control and Flight Recorder Demo Series, among a couple other videos, blog entries, etc.

    We originally tried to use VisualVM but had difficulty getting it running remotely, so we switched to JMC.  When we ran VisualVM locally though, we were able to add filters to the output to get rid of certain classes, allowing us to concentrate on the packages that were directly related to our application's code.  For example we did not profile these packages:

    java.*, javax.*, sun.*, sunw.*, com.sun.*,org.*,groovy.*,groovyjarjarantlr.*,org.*,ice.*, kodo.*,netscape.*,shemacom_bea_xml.*,weblogic.*,commonj.*,oracle.*

    I see that there is a filter field in some areas of mission control, but if we put in our package prefix and a wildcard we only get like 1 result.

    Is there any output that we could provide that would help illuminate this further?  Thanks!

This discussion has been closed.