The following CIMS machine are equipped with Nvidia GPUs and support CUDA development:

  • cuda1.cims.nyu.edu
  • cuda2.cims.nyu.edu
  • cuda3.cims.nyu.edu
  • cuda4.cims.nyu.edu
  • cuda5.cims.nyu.edu

These machines are restricted to logins from within the Courant network, so if you are coming from outside of CIMS, you will have to first login to access.cims.nyu.edu, then use ssh to get to them. Once logged in you can get setup running CUDA code by following these instructions:

To find out what cuda versions are available run:

module avail cuda

Then load the cude version you want to use. For example, to use cuda-10.2 run:

module load cuda-10.2

After loading a cuda module you can copy the sample to your home directory and compile and run them:

cp -r $CUDA_HOME/samples ~/cuda_samples

cd ~/cuda_samples

make

cd bin/x86_64/linux/release

./deviceQuery

./bandwidthTest

See the documentation in $CUDA_HOME/doc/ for more information.

NVIDIA provides online GPU computing seminars.

If you run into problems, please let us know as we'd like to try and get all the kinks worked out.

Thanks go to Professor Petter Kolm for his professor partnership with NVIDIA which provided this hardware.