GTC 2020: AI/ML with vGPU on Openstack or RHV Using Kubernetes
After clicking “Watch Now” you will be prompted to login or join.
Click “Watch Now” to login or join the NVIDIA Developer Program.
AI/ML with vGPU on Openstack or RHV Using Kubernetes
Erwan Gallen, Red Hat | Michael Xiaomin Shen, NVIDIA | Martin Tessun, Red Hat
By sharing GPU resources for AI/ML, you can better utilize on-premises hardware and gain flexibility without moving sensitive workflows into the cloud. Learn how Red Hat Openstack Platform and Red Hat Virtualization are bringing agility to AI/ML accelerated workloads with vGPU. We'll describe how Red Hat contributes to new vGPU capabilities like SR-IOV and live migration support. This makes setting up AI/ML within a Kubernetes system even simpler and more reliable, as the workloads can be migrated and SR-IOV functionality boosts the performance and usability of the vGPU device.