GPU Manager is used for managing the nvidia GPU devices in Kubernetes cluster. It implements the DevicePlugin interface of Kubernetes. So it's compatible with 1.9+ of Kubernetes release version.

To compare with the combination solution of nvidia-docker and nvidia-k8s-plugin, GPU manager will use native runc without modification but nvidia solution does. Besides we also support metrics report without deploying new components.

To schedule a GPU payload correctly, GPU manager should work with gpu-quota-admission which is a kubernetes scheduler plugin.

GPU manager also supports the payload with fraction resource of GPU device such as 0.1 card or 100MiB gpu device memory. If you want this kind feature, please refer to vcuda-controller project.

GPU Manager is running as daemonset, and because of the RABC restriction and hydrid cluster, you need to do the following steps to make this daemonset run correctly.

• service account and clusterrole

kubectl create sa gpu-manager -n kube-system
kubectl create clusterrolebinding gpu-manager-role --clusterrole=cluster-admin --serviceaccount=kube-system:gpu-manager

• label node with nvidia-device-enable=enable

kubectl label node <node> nvidia-device-enable=enable

• change gpu-manager.yaml and submit

change --incluster-mode from false to true, change image field to <your repository>/public/gpu-manager:latest, add serviceAccount filed to gpu-manager-role