[concurrency-interest] ForkJoinPool not designed for nested Java 8 streams.parallel().forEach( ... )
email at christian-fries.de
Wed May 14 16:53:07 EDT 2014
> :-) I think what you have now with CPU burning is more relevant and also different from where you started. Have you tried writing a jmh benchmark instead? i would recommend doing so if you want to further explore the performance characteristics.
I wrote a JMH benchmark, but micro-benchmarking is not the right tool to show the impact of that bug, since the micro-benchmarks will capture „microscopic“ things like thread creation overhead, etc.
whereas the problem appears on the larger scale.
I have created a test case without any blocking operation and the results are that nested parallel loops are a factor 2 slower compared to fixing that bug - see my previous post to concurrency-interest.
For me, the deadlocking is still a more serious problem…
More information about the Concurrency-interest