Using NGC with Azure Setup Guide Table of Contents 1. Introduction to Using NGC with Azure 2. Before You Start 2.1. Prerequisites 2.2. Setting Up SSH Keys 2.3. Setting Up a Security Group 3. Launching an NVIDIA GPU Cloud VM from the Azure Portal 3.1. Creating Your GPU Cloud VM 3.2. Connecting to Your GPU Cloud VM Instance 3.3. Starting and Stopping Your GPU Cloud VM 3.4. Deleting Your GPU Cloud VM and Associated Resources 4. Launching an NVIDIA GPU Cloud VM Using Azure CLI 4.1. Installing Azure CLI 4.2. Preparing Your VM Variables 4.3. Creating Your GPU Cloud VM 4.4. Connecting to Your GPU Instance with SSH 4.5. Stopping (Deallocating) and Starting VMs with the CLI 4.6. Deleting VMs and Associated Resources with the CLI 5. Using Premium Storage SSDs for Datasets 5.1. Creating Data Disk Storage Using the Azure Dashboard 5.2. Creating Data Disk Storage Using the Azure CLI 5.3. Mounting a Data Disk 5.4. Deleting a Data Disk 6. Using File Storage for Datasets 6.1. Creating a Storage Account 6.2. Creating File Storage 6.3. Mounting the File Storage 6.4. Deleting File Storage 7. Examples of Running Containers 7.1. Logging Into the NGC Container Registry 7.2. Example: MNIST Training Run Using TensorFlow Container Notices