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 149108 - Netbeans for multiple cores
Summary: Netbeans for multiple cores
Status: NEW
Alias: None
Product: java
Classification: Unclassified
Component: Source (show other bugs)
Version: 6.x
Hardware: All All
: P3 blocker with 5 votes (vote)
Assignee: Svata Dedic
Depends on:
Reported: 2008-10-03 18:30 UTC by _ gtzabari
Modified: 2013-09-02 14:19 UTC (History)
1 user (show)

See Also:
Exception Reporter:


Note You need to log in before you can comment on or make changes to this bug.
Description _ gtzabari 2008-10-03 18:30:26 UTC
I'm not sure whether this idea is feasible or not (you'd have to measure it) but I noticed that currently Netbeans scans
the classpath using a single core (probably using a single thread). If you were to break this down into multiple threads
then multi-core systems would run the computational end a lot faster. It remains to be seen whether the overall process
is CPU or I/O-bound.

The same technique should probably be checked for the refactoring module ("initializing" takes a long time).
Comment 1 Peter Pis 2008-10-06 10:37:32 UTC
Reassigning to java.
Comment 2 Tomas Zezula 2008-10-06 10:55:18 UTC
It's feasible, it needs to be prototyped. During the initial scan the most load is on the IO, it needs to be measured if adding another core into the game will 
improve the scanning or it will cause IO contention and will slow down the scan process.

Comment 3 pekarna 2012-01-16 01:27:12 UTC
This is still valid request. Project scanning still only happens in a single thread. I/O is not a bottleneck here - I have most files cached and there are barely few reads coming from NetBeans.

Could this be considered for please?
Comment 4 _ gtzabari 2012-12-28 14:37:03 UTC
Another task which takes an inordinate amount of time is code-formatting. This is even more apparent when running on weaker devices like laptops.

Code formatting uses 15% of the CPU (1 out of 8 cores running at 100%) and takes 3-5 seconds on average even for relatively small files.

I think we should convert this into an umbrella issue and evaluate which tasks are best suited for parallelization.