Programming Models for High-Performance Computing

The first version of the MPI standard was released in November 1993. At the time, many of the authors of this standard, myself included, viewed MPI as a temporary solution, to be used until it is replaced by a good programming language for distributed memory systems. Almost twenty years later, MPI is the main programming model for High-Performance Computing, and practically all HPC applications use MPI, which is now in its third generation; nobody expects MPI to disappear in the coming decade. On the other hand, attempts to replace MPI with languages, such as High-Performance-Fortran, failed. Current attempts (UPC, Fortran, X10, Chapel, etc.) face major obstacles and seem unlikely to replace MPI. The talk will discuss some plausible reasons for this situation. These include: 1. Design issues with the various proposed languages: Namely, a few key things that MPI got right and most potential alternatives got wrong 2. Lack of compelling motivation, so far, for switching programming environments 3. The economic and social constraints of High-Performance Computing 4. The large volume of "legacy" MPI code We shall next discuss the implications of this situation for research on new programming models for High-Performance Computing.