Image Information
How to check your software
Running sample code
Tensorflow with GPU test

This page will show you how to test your VM’s GPU and access and test the software that is pre-installed on the scs-gpu-fall-2021-v2 virtual machine.

Image Information

Image Name: scs-gpu-fall-2021-v2
Creation Date: August 16, 2021
Operating System: Ubuntu 20.04
Window Manager: XFCE
Intended usage: Openstack GPU virtual machine with AI programming languages


You will be given a username and password once your VM is ready.

Please change your password as soon as the VM is provisioned for you. This can be done by logging into your VM and then opening a terminal window and typing ‘passwd’.


1. From outside of Carleton you will need to VPN to Carleton in order to access the VM

  • VPN: Carleton VPN when connecting from outside of the campus

2. Listed are the ways you can connect to your VM. x2go is the preferred connection method:

  • x2go: download x2go client to get a  full graphical desktop (use Session Type: XFCE)
  • ssh: use a terminal window to gain ssh access
  • console: SCS Open Stack horizon web interface (please contact SCS Tech Staff)

Recommended usage is to VPN to Carleton and then use the x2go client to connect to your VM


Software Version
NVIDIA Driver 470.57.02
CUDA Driver 11.4
cuDNN 8.2.2
Anaconda 4.10.3 build 3.20.5
GCC 9.3.0
Tensorflow 2.6.0
Pytorch 1.9.0+culll
Keras 2.6.0
Keras-processing 1.1.2
Pandas 1.1.3
Numpy 1.19.2

How to check your software

Check NVIDIA driver version

nvidia-smi -l

Check CUDA version:



nvcc --version

Check cuDNN version:

cat /usr/include/cudnn_version.h

Anaconda version:

conda info

Tensorflow version:

python -c 'import tensorflow as tf; print(tf.__version__)'

Pytorch version:

python -c "import torch; print(torch.__version__)"

Keras version

pip list | grep Keras

Running sample code

CUDA samples are here:


CUDA sample code


cuDNN sample code is located here:


cuDNN sample code


Test Pytorch:

python -c 'import torch;x=torch.rand(5,3);print(x)'

Tensorflow with GPU testing

If you want to test that tensorflow is actually using GPU try running the AI benchmark!

Tensorflow version:

python -c 'import tensorflow as tf; print(tf.__version__)'

Test using Ai benchmark (may take 20 minutes):

python -c 'from ai_benchmark import AIBenchmark;benchmark = AIBenchmark();results ='

Then lookup the results. The chart will tell you how well your GPU is performing. If its not performing well then you may not be using GPU!