CUDA On The CS Linux Clients

CUDA On The CS Linux Clients

CUDA is a parallel computing platform and application programming interface (API) model created by Nvidia.
It will only run on our CS Linux clients that include GPUs.
Although the CS IT Support Group makes the CUDA software available on our GPU-enabled systems, we do not use CUDA ourselves. We therefore do not have the experience necessary to provide instructions on how to use the CUDA toolchain and libraries.
The CUDA tools should work as their standard documentation say they do. If for some reason they do not, please email support@cs.jhu.edu and let us know as much information as possible, and we'll see if we can help correct the issue with our installation.
To learn how to use CUDA, we recommend consulting its online documentation or discussing its use with your instructor.