Computing with GPUs

Updated: June 12, 2020

Edit this Page via GitHub       Comment by Filing an Issue      Have Questions? Ask them here.

Requesting GPUs

GPUs are requested via the GRES option:

sbatch --gres=gpu ...

Request multiple GPUs with:

sbatch --gres=gpu:2

At this time we have one GPU per node.

Nodes with GPUs

Location Partition Node Name GPU
FHCRC largenode j NVIDIA
FHCRC campus-new k NVIDIA
GCP gpu h 1 X Nvidia V100

Using GPUs

When your job is assigned a GPU, Slurm sets the environment variable CUDA_VISIBLE_DEVICES. This environment variable indicates the assigned GPU- most CUDA tools (e.g. tensorflow) use this to restrict execution to that device.

Updated: June 12, 2020

Edit this Page via GitHub       Comment by Filing an Issue      Have Questions? Ask them here.