[concurrency-interest] High performance Future aggregation - polling vs latching

Outside - Karl's ACM kpauls at acm.org
Tue Aug 8 15:33:09 EDT 2006


> -----Original Message-----
> From: Gregg Wonderly [mailto:gregg at cytetech.com]
> Sent: Monday, August 07, 2006 8:33 PM
> 
> How busy will the JVM actually be, how long are these 'waits' and how many
> will
> be running at the same time?  I'm not so sure that the exception
> generation
> overhead (microseconds of CPU perhaps) is going to unduely impact your JVM
> performance.

I'm more concerned about the memory cost.

I did put together a little impl of each method and using the
ExecutorCompletionService is much cleaner than polling and does not lend
itself to implementation errors (such as a poor wait strategy).

I'm considering JavaSpaces for future projects.



More information about the Concurrency-interest mailing list