Providing nested parallelism and load balancing on message passing libraries

A programming methodology and its corresponding tool are presented. The tool expands the message passing library model (PVM, InmosC, MPI, ...), providing nested parallelism, the capacity to mix parallelism and recursion, processor virtualization, load balancing and support for collective operations among other features. Furthermore, as it is a valuable tool for the presentation and teaching of parallel algorithms, the computational results show an efficiency similar to that obtained using the native language