Using NGC with Google Cloud Platform Setup Guide Table of Contents 1. Introduction to Using NGC with Google Cloud Platform 2. Before You Start 2.1. Prerequisites 2.2. About Shared Projects 2.3. Setting Up SSH Keys 2.4. Setting Firewall Rules 3. Deploying an NVIDIA GPU Cloud Image from the GCP Console 3.1. Creating Your GPU Instance 3.2. Connecting to Your GPU Instance 3.3. Stopping and Restarting Your VM Instance 4. Launching an NVIDIA GPU Cloud Image with the gcloud CLI 4.1. Installing and Setting Up gcloud CLI 4.2. Preparing the Create Instance Options 4.3. Creating a Custom GPU Instance 4.4. Connecting to Your GPU Instance with SSH 4.5. Stopping and Restarting Your GPU Instance 4.6. More Advanced CLI Usage 5. Creating Persistent SSD Disks for Datasets 5.1. Creating a Persistent SSD Dataset Disk and Attaching it to the VM Instance from the GCP Console 5.2. Creating a Persistent SSD Dataset Disk and Attaching it to the VM Instance from the gcloud CLI 5.3. Creating a File System and Mounting the Volume 5.4. Deleting a Persistent SSD Disk 6. Examples of Running Containers 6.1. Logging Into the NGC Container Registry 6.2. Example: MNIST Training Run Using PyTorch Container 6.3. Example: MNIST Training Run Using TensorFlow Container 6.4. Example: Persistent SSD Dataset Disk with ImageNet, ResNet50 Tor TensorFlow Notices