Once we have the main array broken, we send list1 and list2 to the ForkJoinSumArrayTask, then we fork the task1, which executes the compute() method and the other part of the array in parallel.

A team of researchers at the University of Illinois' Universal Parallel Computing Research Center (UPCRC) has developed a parallel implementation of Java.

Java integers are stored in 32 bits, so each vector register can hold eight Java integers (ints). The JIT will generate code to load values from the array in groups of eight.

Native support for HSA’s specifications in JVMs is expected with Java 9 in 2015, and parallel algorithms will be executed natively in JVMs without extra layers of code.