Last updated 11th December 2018
The purpose of this guide is to show you how to set up your first NVIDIA GPU Cloud container instance.
The first part of the guide will be OVH specific for basic usage. The second part will be more advanced, using OpenStack client to automate NGC instance management.
- Access to the OVH Control Panel.
Create an NGC instance via OVH Manager
The first step will be to create a Public Cloud project
The second step will be to create an instance into the newly created project.
In the case of NGC, be sure to pick:
NVIDIA GPU Cloud (NGC)as your image
- a GPU enabled flavor like
t1-90for 1 or 2 NVIDIA® Tesla® V100.
Create an NGC instance via the Command Line Interface
You can manage your OVH Public Cloud instances using the standard OpenStack API and tools:
For now, we'll focus on the command line client
Configure your local environment
Next, click on the
... icon at the end of the line and then click the
Download OpenStack configuration file link.
Save the file as
If you are using Windows, please follow this link to setup the OpenStack client:
If you are using a Unix-based OS, use your preferred package manager (i.e.
emerge, etc) to install the
python-openstackclient package and source the configuration file you previously saved (
Create a NGC VM
The first step is to have a SSH Key pair. You can create one with:
openstack keypair create mykey > mykey.pem
Then, you'll need to gather the following information:
- the SOURCE ID:
openstack image list --name 'NVIDIA GPU Cloud (NGC)'
- the FLAVOR:
- the NETWORK ID:
openstack network list --name 'Ext-Net'
Finally, create the VM with the following command:
openstack server create --key-name mykey --image $SOURCE_ID --flavor $FLAVOR --network $NETWORK_ID my_vm
The billing starts as soon as the virtual machine is up.
You can get the VM status with
openstack server show my_vm and delete it with
openstack server delete my_vm.
SSH to the VM
Get the VM IP with
openstack server show my_vm and SSH into it with
ssh -i ./mykey.pem ubuntu@<VM IP>
Your first NGC container
Once you are logged into the VM, you can start pulling and running the container.
The list of available containers (TensorFlow, Caffe2, DIGITS, Matab, MXNet, PyTorch, TensorFlow, RAPIDS, ...) is available here:
Please click the link below to see an example of semantic segmentation with NVIDIA DIGITS:
docker pull nvcr.io/nvidia/digits:18.11-tensorflow docker run -p 8888:5000 nvcr.io/nvidia/digits:18.11-tensorflow
Then navigate to
Please click the link below for more options and explanations:
If you would like to automated the above steps, look at the
-f json option of the openstack client and the
jq JSON command line tool parser.
SOURCE_ID=`openstack image list --name 'NVIDIA GPU Cloud (NGC)' -f json | jq -r '..ID'`
Join our community of users on https://community.ovh.com/en/.