System requirements

​

System configuration

• Operating System: Ubuntu LTS 20/22/24
• X86 architecture
• RAM: 64G
• Disk space: 256Gb + 2.5 times the estimated amount of data.
• CPU: 16 cores
• Optional: GPU: Common Nvidia GPU types (Nvidia A10, A100, H100, RTX 4090, etc.)

​

Dependencies

• Docker v.27 or above - follow the official instructions and the post-install guide.
• jq - json data parsing library
• psql - postgres command line client
• zip/unzip

​

NVIDIA GPU Setup (GPU machines only)

1. Add NVIDIA package repositories:
   Copy

   Ask AI

   curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg
   curl -fsSL 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://#' \
     | sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list
   

2. Download the nvidia-container-toolkit:
   Copy

   Ask AI

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

3. Configure Docker to use the NVIDIA runtime:
   Copy

   Ask AI

   sudo nvidia-ctk runtime configure --runtime=docker
   sudo systemctl restart docker
   

4. Verify the installation:
   Copy

   Ask AI

   docker run --rm --runtime=nvidia --gpus all nvidia/cuda:11.5-base-ubuntu20.04 nvidia-smi
   

   This should display your GPU information if the toolkit is properly installed.

​

Advised tools for troubleshooting/maintenance

• vi - or equivalent required for file editing
• object store CLIs if applicable (aws cli, mc)

​

Required permissions and access

• Make sure you run the installation as a regular user (not a root user), with sudo privileges.