NVidia GPU support in Red Hat OpenShift Container Platform

Solution In Progress - Updated -

Issue

  • One of our development team wants to integrate nvidia GPUs for high performance computing. Therefore they need a host system with nvidia GPUs and some nvidia drivers (nvidia cuda environment). It would be great to integrate those hosts as openshift nodes. But to use them we must launch containers with the following parameters: docker run -ti --rm --volume-driver=nvidia-docker --volume=nvidia_driver_367.48:/usr/local/nvidia:ro \
    --device=/dev/nvidiactl --device=/dev/nvidia-uvm --device=/dev/nvidia-uvm-tools --device=/dev/nvidia0 \
    --device=/dev/nvidia1 --device=/dev/nvidia2 --device=/dev/nvidia3 nvidia/cuda nvidia-smi
    . Is it possible to start containers (pods) with special volume drivers, volumes and devices with openshift?

Environment

  • Red Hat Container Platform 3.2.1

Subscriber exclusive content

A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.

Current Customers and Partners

Log in for full access

Log In
Close

Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.