CUDA

CUDA, created by NVIDIA, is a model and platform for parallel computing on GPU (Graphics Processing Units). CUDA runs on CUDA-Capable GPUs which are created by NVIDIA. It was built to be used with C, C++ and Fortran but there are third-party wrappers available for all kinds of other languages including, but not limited to Python, Ruby, Perl. Two of our clusters have NVIDIA graphics cards -- Al-Salam has Tesla cards and Layout has GeForce GTX cards. CUDA is installed from source as an environment module. Please read Environment Modules before proceeding to the install process.