Self-Hosted installation over Kubernetes - Prerequisites¶

Before proceeding with this document, please review the installation types documentation to understand the difference between air-gapped and connected installations.

Run:ai Components¶

As part of the installation process you will install:

A control-plane managing cluster
One or more clusters

Both the control plane and clusters require Kubernetes. Typically the control plane and first cluster are installed on the same Kubernetes cluster but this is not a must.

Installer machine¶

The machine running the installation script (typically the Kubernetes master) must have:

At least 50GB of free space.
Docker installed.

Helm¶

Run:ai requires Helm 3.14 or later. To install Helm, see Installing Helm. If you are installing an air-gapped version of Run:ai, The Run:ai tar file contains the helm binary.

Cluster hardware requirements¶

See Cluster prerequisites hardware requirements.

In addition, the control plane installation of Run:ai requires the configuration of Kubernetes Persistent Volumes of a total size of 110GB.

Run:ai software requirements¶

Cluster Nodes¶

See Run:ai Cluster prerequisites operating system requirements.

Nodes are required to be synchronized by time using NTP (Network Time Protocol) for proper system functionality.

Kubernetes¶

See Run:ai Cluster prerequisites Kubernetes distribution requirements.

The Run:ai control plane operating system prerequisites are identical.

The Run:ai control-plane requires a default storage class to create persistent volume claims for Run:ai storage. The storage class, as per Kubernetes standards, controls the reclaim behavior: whether the Run:ai persistent data is saved or deleted when the Run:ai control plane is deleted.

Note

For a simple (nonproduction) storage class example see Kubernetes Local Storage Class. The storage class will set the directory /opt/local-path-provisioner to be used across all nodes as the path for provisioning persistent volumes.

Then set the new storage class as default:

kubectl patch storageclass local-path -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'

Install prerequisites¶

Ingress Controller¶

The Run:ai control plane installation assumes an existing installation of NGINX as the ingress controller. You can follow the Run:ai Cluster prerequisites Kubernetes ingress controller installation.

NVIDIA GPU Operator¶

See Run:ai Cluster prerequisites NVIDIA GPU operator requirements.

The Run:ai control plane, when installed without a Run:ai cluster, does not require the NVIDIA prerequisites.

Prometheus¶

See Run:ai Cluster prerequisites Prometheus requirements.

The Run:ai control plane, when installed without a Run:ai cluster, does not require the Prometheus prerequisites.

Inference (optional)¶

See Run:ai Cluster prerequisites Inference requirements.

The Run:ai control plane, when installed without a Run:ai cluster, does not require the Inference prerequisites.

External Postgres database (optional)¶

The Run:ai control plane installation includes a default PostgreSQL database. However, you may opt to use an existing PostgreSQL database if you have specific requirements or preferences. Please ensure that your PostgreSQL database is version 16 or higher.

Next steps¶

Continue to Preparing for a Run:ai Kubernetes Installation .