[concurrency-interest] Fixed thread pool problem?

Doug Lea dl at cs.oswego.edu
Fri Aug 31 19:15:30 EDT 2007


Glyn Normington wrote:
> so the 
> only source code I've found for the concurrency utilities is some early 
> "public domain" code, so I haven't been able to get much further in 
> analysis.
> 

(Not that you needed see them to answer this,
but the definitive public domain sources
for j.u.c. are always in our CVS -- see the usual links at
http://gee.cs.oswego.edu/dl/concurrency-interest/index.html --
although we don't always know exactly which CVS revisions
any particular version of any JVM is based on.)

> and watch the CPU utilisation using a system monitor. On a dual or quad 
> CPU system, this stays at around one processor's worth.
> 

I think the lack of speedup is mostly due to synchronization
in Generation.equals? It seems that there's a lot of
lock contention there. (This is in addition to the explanation
of the infinite loop anomalies by Joe.)

(BTW, For fun, someone might want to try a ParallelArray version
of this code?)

-Doug


More information about the Concurrency-interest mailing list