I'm pleased to report that the recently released v22.9.1 of the #NVIDIA #GPU Operator from #NGC https://catalog.ngc.nvidia.com/orgs/nvidia/helm-charts/gpu-operator, installs and runs perfectly on #Kubernetes 1.26 #Electrifying. Tested on dev cluster with #TensorRT.

To keep #docker container maintenance to a minimum, instead of creating one for each job, we now NFS mount the #Python source folder, with subPath, and pull only generic containers. We mostly use latest image, but version specific jobs still use tags, e.g. running #TensorRT #inference models.