Linux

Everything you need to know about deploying Supervisely agent on Linux based operating systems

Deploy Supervisely agent on Linux

Supervisely agent can work both with and without GPU support. If you don't have a GPU, you can deploy the agent on any machine with Linux OS and you can skip the GPU installation steps. This tutorial explains how to deploy the Supervisely agent on Linux OS.

Table of Contents

• Prerequisites

• How to install

• Step 1. Install Docker

• Step 2. Install CUDA Toolkit

• Step 3. Install NVIDIA Driver

• Step 4. Install NVIDIA Container Toolkit

• Step 5. Deploy Supervisely Agent

• Troubleshooting

Prerequisites

• Linux OS (Kernel 3.10 or higher)

• Docker (Version 19.3 or higher)

• CUDA Toolkit (Version 9.0 or higher)

• NVIDIA Driver (Version 452.39 or higher)

• NVIDIA Container Toolkit

How to install

Step 1. Install Docker

First, set up the Docker apt repository:

# Add Docker's official GPG key:
sudo apt-get update
sudo apt-get install ca-certificates curl
sudo install -m 0755 -d /etc/apt/keyrings
sudo curl -fsSL https://download.docker.com/linux/ubuntu/gpg -o /etc/apt/keyrings/docker.asc
sudo chmod a+r /etc/apt/keyrings/docker.asc

# Add the repository to Apt sources:
echo \
  "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/ubuntu \
  $(. /etc/os-release && echo "$VERSION_CODENAME") stable" | \
  sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
sudo apt-get update

Then install the Docker packages:

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

Finally, verify that the Docker Engine installation is successful by running the hello-world image:

 sudo docker run hello-world

Check out the official Docker documentation for more information.

Step 2. Install CUDA Toolkit

Install the CUDA Toolkit using the following commands:

wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-ubuntu2204.pin
sudo mv cuda-ubuntu2204.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com/compute/cuda/12.4.1/local_installers/cuda-repo-ubuntu2204-12-4-local_12.4.1-550.54.15-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu2204-12-4-local_12.4.1-550.54.15-1_amd64.deb
sudo cp /var/cuda-repo-ubuntu2204-12-4-local/cuda-*-keyring.gpg /usr/share/keyrings/
sudo apt-get update
sudo apt-get -y install cuda-toolkit-12-4

Check out the official CUDA Toolkit documentation for more information and the latest version.

Step 3. Install NVIDIA Driver

Install the NVIDIA driver using the following commands:

sudo apt-get install -y cuda-drivers-550

To verify the installation, run the following command:

nvidia-smi

It's recommended to restart your system after installing the NVIDIA driver with the following command:

sudo reboot

The output should display the NVIDIA driver version, CUDA version, and GPU information.

nvidia-smi

If you can see this information, the installation was successful. Otherwise, please check the Troubleshooting section.

Check out the official NVIDIA Driver documentation for more information and the latest version.

Step 4. Install NVIDIA Container Toolkit

The NVIDIA drivers must be also available in the Docker containers so the agent can utilize the GPU. To do this, install the NVIDIA Container Toolkit using the following commands:

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/stable/deb/nvidia-container-toolkit.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
sudo apt-get update
sudo apt-get install -y nvidia-container-toolkit

Now, configure the Docker daemon to use the NVIDIA runtime:

sudo nvidia-ctk runtime configure --runtime=docker

Finally, restart the Docker daemon:

sudo systemctl restart docker

Now, we'll need to ensure that the NVIDIA Container Toolkit is installed and working correctly and that the NVIDIA runtime is available inside the Docker containers. To do this, run the following command:

sudo docker run --rm --runtime=nvidia --gpus all nvidia/cuda:12.4.1-base-ubuntu22.04 nvidia-smi

The output should display the NVIDIA driver version, CUDA version, and GPU information.

nvidia-smi in Docker

If you can't see this information, please check the Troubleshooting section.

Step 5. Deploy Supervisely Agent

Now it's time to deploy the Supervisely agent.

Open Supervisely instance, go to the Cluster page and press the Add button. Select the Supervisely agent option.

Add Agent

Copy the command and run it in the terminal on the machine where you want to deploy the agent.

Command

That's it! Now your agent is deployed and running.

TroubleShooting

If the nvidia-smi command does not display the GPU information from OS or the Docker container, the NVIDIA drivers were not successfully installed. In this case, you can try to uninstall the NVIDIA drivers and CUDA Toolkit:

sudo apt-get purge nvidia*
sudo apt-get purge cuda*
sudo apt-get autoremove
sudo apt-get autoclean

After that restart your system, and try to install the components again.