This Bugzilla instance is a read-only archive of historic NetBeans bug reports. To report a bug in NetBeans please follow the project's instructions for reporting issues.

Bug 55221 - Merged CPU results
Summary: Merged CPU results
Alias: None
Product: profiler
Classification: Unclassified
Component: Base (show other bugs)
Version: 4.x
Hardware: PC All
: P3 blocker with 5 votes (vote)
Assignee: issues@profiler
: 117409 (view as bug list)
Depends on:
Reported: 2005-02-18 17:43 UTC by iformanek
Modified: 2016-11-09 13:02 UTC (History)
3 users (show)

See Also:
Exception Reporter:


Note You need to log in before you can comment on or make changes to this bug.
Description iformanek 2005-02-18 17:43:17 UTC
Provide a mode of CPU results, where the results 
from all threads are merged into a single tree.

This is aprticularly useful for people that do 
not care about the threads in their app (which 
could be majority), and app/web servers where 
there could be a pool of worker threads, and the 
user does not necessarily care which of them 
executes this code.
Comment 1 iformanek 2005-03-29 15:01:15 UTC
I believe this is planned for 1.0, right?
Comment 2 mishadmitriev 2005-03-31 04:22:04 UTC
Yes, if time permits (not a defining or target feature).
Comment 3 iformanek 2005-06-20 11:07:11 UTC
Not for 1.0
Comment 4 ghirschhorn 2005-07-12 19:18:48 UTC
This is also useful cases where the same work is distributed over many threads.
In many of these cases, only an aggregate tree can show the high-level method
that is the cause of a performance problem, or at least the starting point for
digging further.  
Comment 5 Alexander Kouznetsov 2007-10-02 10:03:17 UTC
*** Issue 117409 has been marked as a duplicate of this issue. ***
Comment 6 Alexander Kouznetsov 2007-10-02 10:03:38 UTC
From issue 117409:

Feedback from user: I still have the same problem in Netbeans 5.5.1:  I am trying to optimize a web application that
runs in a number of worker threads and that makes calls to back-end servers such as data base or EJB servers.  The
following feature would help a lot:
     A Call Tree view that shows an aggregate over all the worker
     threads in my server.  This would give me a chance to find the
     methods that consume the most time by top-down opening the call tree.

As it is, this deficiency conspire to make it difficult to find the methods that actually take the most total time: the
Call Tree view is spread out over many threads that I must inspect individually to locate the worst method.

Apart from this, I found the profiler easy to work with (especially compared to Eclipse TPTP, which I have used before
and which insists on recording every single function call instead of an aggregate).
Comment 7 Tomas Hurka 2009-04-08 12:55:24 UTC
Milestone cleanup: future->next
Comment 8 Jiri Sedlacek 2016-11-09 13:02:46 UTC
Implemented in the new profiler.