A simple guide to install Docker and Nvidia Docker to get CUDA Support on Docker Images

Build and Run NVIDIA Docker with CUDA 11.3 on Pop!_OS or Ubuntu 20.04 LTS

#!/bin/bash

### Step 1: Install Docker ###

    curl -fsSL https://get.docker.com -o get-docker.sh

    sudo sh get-docker.sh
    

### Step 2: Restart the Docker ###

    sudo systemctl restart docker
    

### Step 3: Add the package repositories for Nvidia-Docker ###

    distribution=ubuntu20.04

    curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
    
    curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list


### Step 4: Install Nvidia-Docker ###

    sudo apt-get update && sudo apt-get install -y nvidia-container-toolkit

    sudo systemctl restart docker


### Step 5: Get CUDA 11.3 for Nvidia-Docker ###

    docker pull nvidia/cuda:11.3.0-base-ubuntu20.04


### Step 6: Run Nvidia Docker with CUDA ###

    docker run --rm --gpus all nvidia/cuda:11.3.0-base-ubuntu20.04 nvidia-smi


### Step 7: Add host machine local storage to docker container ###

    docker container run -v /home/faisal/workspace/:/home/ --gpus all -it <IMAGE_ID> bin/bash