../common/generic-attributes.adoc = Preparing the cluster for {ailibrary} This procedure assumes that you already have the base operating system installed on cluster nodes as well as the {rke2} {kube} distribution installed and operational. If you are installing from scratch, refer to [ai-deployment-kube-installing] first.
+ . Install the {nvoperator} as described in [nvidia-operator-installation]. . Connect the downstream {kube} cluster to {ranchermanager} running on the upstream cluster as described in [rancher-register-clusters]. . Configure the GPU-enabled nodes so that the {productname} containers are assigned to Pods that run on nodes equipped with {nvidia} GPU hardware. Find more details about assigning Pods to nodes in [ai-gpu-nodes-assigning]. . (Optional) Install {ssecurity} as described in [ai-security-installation]. Although this step is not required, we strongly encourage it to ensure data security in the production environment. . Install and configure {sobservability} to observe the nodes used for {productname} application. Refer to [observability-installing] for more details.