Kubernetes on NVIDIA GPUs

Scale-out GPU-Accelerated Applications

Kubernetes on NVIDIA GPUs

Kubernetes on NVIDIA GPUs enables enterprises to scale up training and inference deployment to multi-cloud GPU clusters seamlessly. NVIDIA is developing GPU enhancements to Kubernetes and working closely with the Kubernetes open-source community to contribute GPU enhancements for the benefit of the larger ecosystem.

With Kubernetes on NVIDIA GPUs, developers can deploy GPU-accelerated deep learning and HPC applications to multi-cloud GPU clusters instantly.

Key capabilities include:

  • Enables support for GPUs in Kubernetes using the NVIDIA device plugin
  • Deploy applications to a heterogeneous GPU cluster in the cloud or datacenter
  • Maximize GPU utilization using GPU sharing
  • Monitor health and metrics from GPUs in Kubernetes using datacenter management tools such as NVML and DCGM

Please sign up for the interest list below to be notified when it is available.

(Click to Zoom)


NVIDIA Container Runtime

NVIDIA Container Runtime simplifies the process of building and deploying containerized GPU-accelerated applications to desktop, cloud or data centers. Developers can wrap their GPU-accelerated applications along with its dependencies into a single package that is guaranteed to deliver the best performance on NVIDIA GPUs, regardless of the deployment environment.

With NVIDIA Container Runtime, researchers and application developers in deep learning and HPC can deploy and run their GPU-accelerated applications using different container technologies such as Docker or LXC.

NVIDIA Container Runtime for Docker

NVIDIA Container Runtime for Docker is an open-source project hosted on GitHub. NVIDIA GPU Cloud documentation has more information on how to install and get started with Docker containers for deep learning frameworks.


NVIDIA Container Runtime also supports LXC, which is available on different Linux distributions.