Data Parallelism with the Fork-Join Framework in Java 7

Data Parallelism with the Fork-Join Framework in Java 7

The fork/join framework added to the Java SE 7 release is an implementation of the ExecutorService interface that helps you take advantage of multiple processors. Although concurrent programming on multiprocessors has existed for decades, the evolution of multicore is bringing true parallelism to nearly every computer in your life. To take full advantage in a multicore world, software needs to be parallelized and it needs to scale. One approach to data parallelism is through divide and conquer techniques with which problems are broken into chunks that can be solved in parallel and recombined. Employing these techniques requires library support, and in Java 7, this is brought to you via the java.util.concurrent fork/join framework. This session explores how to use fork/join to construct parallel solutions to common problems such as parallel sorting.