Computer Services Calendar Safety     Search PSFC



PSFC GPU Cluster


The cluster consists of six nodes with four nvidia gpus each.


Three nodes have four v100s cards and three nodes have four rtx6000 cards. The former are double precision and the latter are more appropriate for machine learning like workflows.

Requesting Access

To request access to the PSFC GPU cluster, send an email to: with your name and PSFC username.

Software Stack

CentOS 8.2
Python 3
CUDA 10.2 (nvcc)
Tensorflow 2 with GPU support
OpenMPI 4.0.4
PGI CUDA Fortran compiler (nvfortran)

Accessing the Cluster

You need to be on the PSFC network in order to access the cluster. If you are not the network, you can connect to it using the PSFC VPN. Please follow the instructions at


Log into the cluster using ssh to You can also access this node using the OnDemand web interace This is the gateway node and you launch your jobs from here. Note that does not have nvidia cards or compilers. You have to request an interactive job to access a node for compilation. You can also currently login directly to one of the gpu nodes as well, but please only use that for quick compilations not running jobs.


The v100s nodes are:, and


The RTX6000 nodes are:, and

Job Management

SLURM is used for job management just as on engaging. An example job requesting one gpu card per node on 2 nodes:


   #SBATCH --job-name=hello_world
   #SBATCH --output=hello.txt
   #SBATCH --ntasks-per-node=1
   #SBATCH --nodes=2
   #SBATCH --gres=gpu:1
   #SBATCH --time=05:00
   #SBATCH -p regular

   mpirun hostname
   mpirun nvidia-smi

Help Desk for the PSFC GPU Cluster, email:


Note: current documentation is non-existent, so please address any questions you have on usage to:


77 Massachusetts Avenue, NW16, Cambridge, MA 02139,


massachusetts institute of technology