Nvidia has partnered with three U.S. Energy Department-funded national laboratories and the National Nuclear Security Administration to build an open source compiler technology for high-performance computing systems.
Nvidia’s PGI software business, NNSA and the Los Alamos, Sandia and Lawrence Livermore national laboratories will collaborate on the development of a Fortran-based architecture that the alliance intends to integrate with the LLVM software compilation framework, the Lawrence Livermore National Laboratory said Nov. 13.
The department launched the multiyear public-private initiative as it seeks a common compiler platform to support Clang C++ and Fortran applications.
Organizations use the Fortran programming language in the simulation, national security, energy, basic science, climate and medical research areas, Lawrence Livermore noted.
PGI, NNSA and the three national labs aim to make an open-source Fortran compiler front-end module available for public access by late 2016.