Please login through the login node first, reserve an interactive session in Slurm, then should be able to ssh to the GPU node.
Login to node using ssh netid@dkucc-login-01.oit.duke.edu
Then run sinfo to get a list of partition and nodelist names, as below.
$ sinfo
PARTITION AVAIL TIMELIMIT NODES STATE NODELIST
common* up 90-00:00:0 1 mix dkucc-core-01
common* up 90-00:00:0 7 idle dkucc-core-[02-08]
common-gpu up 7-00:00:00 1 alloc dkucc-core-gpu-01
common-gpu up 7-00:00:00 1 idle dkucc-core-gpu-02
The interactive session for a GPU node can be done on the DKUCC with the command:
srun -p common-gpu --gres=gpu:1 --exclusive --pty bash -i
