Parallel computation framework for Java

 Java framework that supports a style of parallel programming
 in which problems are solved by (recursively) splitting them
 into subtasks that are solved in parallel, waiting for them
 to complete, and then composing results.
 .
 Fork/Join parallelism is among the simplest and most
 effective design techniques for obtaining good parallel
 performance. Fork/Join algorithms are parallel versions
 of familiar divide-and-conquer algorithms.
 .
 JSR166y includes a parallel computation framework:
 ForkJoinTasks and their related support classes provide a
 very efficient basis for obtaining platform-independent
 parallel speed-ups of computation-intensive operations.