SIO Setup Tutorial (Docker)

This document walks through the steps required to run SIO Docker image on Ubuntu

Step 1: Ubuntu Upgrades

Start with updating and upgrading Ubuntu packages
```
sudo apt-get update
sudo apt-get upgrade
```

Step 2: NVIDIA GPU Drivers Installation

Try the following command before continuing with the GPU driver installation instructions to see if they're already installed on the system:
```
nvidia-smi
```
If the command worked it will show an output similar to the following. If so, skip to Step 3.
Follow the instructions on the driver web page: Advanced Driver Search official NVIDIA drivers.
Below is an example taken from installation instructions for a RTX 3060 GPU.
- Once you have downloaded the driver, change to the directory containing the driver package and install the driver by running, as root,
```
sudo sh ./NVIDIA-Linux-x86_64-525.60.11.run
```
Reboot system
```
sudo reboot
```
After the reboot, make sure the GPU drivers were successfully installed by running the following command:
```
nvidia-smi
```

Step 3: Docker Installation

Install the Docker Engine by following the instructions on Docker's website, or use the following commands. If you run into any issues, Docker's page may have troubleshooting info that can help.

Uninstall any previous installations

sudo apt-get remove docker docker-engine docker.io containerd runc

Update and install required packages for Ubuntu

sudo apt-get update
sudo apt-get install \
    ca-certificates \
    curl \
    gnupg \
    lsb-release

Add Docker GPG key

sudo mkdir -p /etc/apt/keyrings
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o /etc/apt/keyrings/docker.gpg

Repository setup

echo \
  "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu \
  $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

Install Docker

sudo apt-get update
sudo apt-get install docker-ce docker-ce-cli containerd.io docker-compose-plugin

To verify if the installation worked try the following

sudo docker run hello-world
sudo docker compose version

To use Docker commands without sudo
```
sudo usermod -aG docker $USER
```
The terminal needs to be restarted for this to take effect, I prefer a reboot
```
sudo reboot
```
Verify you can run commands without sudo
```
docker ps
```

Step 4: NVIDIA Container Toolkit

This allows Docker to use the installed GPU. Full documentation can be found on NVIDIA's Installation Guide.

Setup package repository and GPG key

distribution=$(. /etc/os-release;echo $ID$VERSION_ID) \
      && curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
      && curl -s -L https://nvidia.github.io/libnvidia-container/$distribution/libnvidia-container.list | \
            sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
            sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

Install nvidia-docker2

sudo apt-get update
sudo apt-get install -y nvidia-docker2

Restart Docker
```
sudo systemctl restart docker
```

To verify if the GPU can be used with Docker

sudo docker run --rm --gpus all nvidia/cuda:11.0.3-base-ubuntu20.04 nvidia-smi

Step 5: Get SIO Docker Image

The SIO docker image can be obtained from our private registry, once we've created an account for you and provided you with sighthound-keyfile.json:

cat sighthound-keyfile.json | docker login -u _json_key --password-stdin us-central1-docker.pkg.dev
docker pull us-central1-docker.pkg.dev/ext-edge-analytics/docker/sio:r240409

At the end of the command above, r240409 represents the release that can be changed depending on the current release version. Latest release as of April 9, 2024 was r240409.
More details are available at Working with SIO framework.

Step 6: Run SIO pipelines

Here’s an example that uses the GPU and runs the folder watcher pipeline with the Docker image that we pulled in Step 4.

docker run -it --gpus all --rm \
-v /data:/data \
-e SIO_DATA_DIR=/data \
us-central1-docker.pkg.dev/ext-edge-analytics/docker/sio:r240409 \
/sighthound/sio/bin/runPipeline \
/sighthound/sio/share/pipelines/VehicleAnalytics/VehicleAnalyticsFolderWatch.yaml \
folderPath=/data/watchedFolder \
--license-path /data/sighthound-license.json

More details are available at Working with SIO framework.