nvidia-docker
A deprecated wrapper that enabled Docker containers to access NVIDIA GPU resources.
What is nvidia-docker?
nvidia-docker was a wrapper tool that enabled Docker containers to access and utilize NVIDIA GPU hardware. It solved the problem of running GPU-accelerated applications, such as machine learning models or scientific simulations, within isolated container environments by providing the necessary runtime configuration.
Developers and researchers working with GPU-accelerated applications in containerized environments, particularly in fields like machine learning, deep learning, and high-performance computing.
It simplified the complex setup required to expose GPU resources to Docker containers, making GPU workloads more portable and easier to deploy compared to manual configuration. However, it has been deprecated in favor of the more comprehensive NVIDIA Container Toolkit.
Overview
Build and run Docker containers leveraging NVIDIA GPUs
Use Cases
Best For
- Running deep learning frameworks like TensorFlow or PyTorch in Docker containers
- Deploying GPU-accelerated scientific computing applications in isolated environments
- Developing and testing machine learning models with consistent GPU access across systems
- Setting up reproducible research environments that require NVIDIA GPU resources
Not Ideal For
- New projects starting today that require NVIDIA GPU support in Docker containers
- Production systems where long-term maintenance, security updates, and compatibility with latest Docker versions are essential
- Environments needing advanced features like multi-GPU management or cloud-native integrations beyond basic GPU access
Pros & Cons
Pros
Provided a wrapper around Docker commands to expose NVIDIA GPUs to containers without complex manual configuration, as described in the README for simplifying GPU workload deployment.
Enabled consistent GPU access across different systems, making it easier to run machine learning and scientific applications in isolated containers, promoting reproducibility.
Handled necessary runtime settings to ensure GPU devices and libraries were available inside containers, reducing setup overhead for developers.
Cons
The project has been archived and replaced by the NVIDIA Container Toolkit, meaning no updates, bug fixes, or official support, as stated in the README's deprecation notice.
Exclusively designed for NVIDIA GPUs, so it's unsuitable for environments with other GPU vendors like AMD or Intel, limiting flexibility.
Relied on a custom Docker wrapper that could introduce compatibility issues with newer Docker versions or alternative container runtimes, adding fragility.
Frequently Asked Questions
Related Projects
ansibleAnsible is a radically simple IT automation platform that makes your applications and systems easier to deploy and maintain. Automate everything from code deployment to network configuration to cloud management, in a language that approaches plain English, using SSH, with no agents to install on remote systems. https://docs.ansible.com.
localstack💻 A fully functional local AWS cloud stack. Develop and test your cloud & Serverless apps offline
traefikThe Cloud Native Application Proxy
PortainerMaking Docker and Kubernetes management easy.
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.