Thanks Michael for the feedback! I have asked to some Akka core dev and it seems that even if the tasks aren't optimized for divide and conquer the FJ pool is able to perform work-stealing without blocking under high contention and high load, differently from the Artemis pool that is backed by a BlockingQueue. That would improve the scaling anyway. The only drawback seems related to blocking operations and on Artemis we have a lot (java and intrinsic locks mainly): they should need to be managed using https://docs.oracle.com/javase/8/docs/api/java/util/concurrent/ForkJoinPool.html#managedBlock-java.util.concurrent.ForkJoinPool.ManagedBlocker-
Do you have any idea on how to use it? -- Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-Dev-f2368404.html