Kubernetes Installation#
NVIDIA Mission Control uses separate Kubernetes clusters for administrative services and user workloads. Run the BCM setup tools on the active BCM head node to deploy these clusters onto their Kubernetes nodes. The BCM head node is distinct from the Kubernetes admin and user control plane nodes.
The cluster and node category names used in these instructions are defined in the Kubernetes installation naming reference.
Administrative Cluster (k8s-admin)#
The k8s-admin control plane is created during initial deployment. This cluster is the primary management layer. It hosts core infrastructure services and operational components, including:
Autonomous Job Recovery
Autonomous Hardware Recovery
Dynamic Power Software
Observability stack
User Cluster (k8s-user)#
The k8s-user cluster hosts user GPU workloads. Its Kubernetes control plane
runs on the k8s-system-user nodes. Both clusters are deployed in
Staging and Installing Artifacts. After that deployment and the
preparation steps in Overview and Prerequisites for NMC Components, install Run:ai onto
k8s-user as described in Run:ai Installation.
The administrative cluster does not replace the user cluster.