> ## Documentation Index
> Fetch the complete documentation index at: https://docs.rafftechnologies.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Kubernetes

> Managed Kubernetes clusters — free single-master control plane, optional paid HA, autoscaling node pools, billed per worker node

<sub>Updated August 24, 2026</sub>

Raff Kubernetes is a managed Kubernetes service. Pick a worker plan and a node count; 10–20 minutes later you have a running cluster and an admin kubeconfig. We operate the control plane, keep it monitored, and roll out version upgrades in place — you deploy with the tools you already use: kubectl, Helm, your CI.

The control plane is free — you pay only for worker nodes, from \$9.99/month each. High availability (3 masters and a redundant gateway) is a flat \$30/month. Every cluster runs RKE2 with Cilium networking in its own VPC, behind a managed gateway with a dedicated public IP.

<Frame>
  <img src="https://mintcdn.com/rafftechnologiesllc/pydLUQqixwIwkrCw/images/products/compute/kubernetes/hero.png?fit=max&auto=format&n=pydLUQqixwIwkrCw&q=85&s=eb4e8c36333b7f0cec5b91a7bbcba93b" alt="Kubernetes cluster overview in the Raff dashboard — live node, pod, CPU, memory and control-plane health cards" width="1651" height="992" data-path="images/products/compute/kubernetes/hero.png" />
</Frame>

## At a glance

| Property                    | Value                                                                                                                                              |
| --------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Kubernetes distribution** | RKE2 — v1.33 to v1.35 — see [Supported versions](/products/build/kubernetes/details/supported-versions)                                            |
| **Control plane**           | Managed and free; optional [HA (3 masters)](/products/build/kubernetes/concepts/control-plane-and-ha) for \$30/month                               |
| **Worker nodes**            | [Node pools](/products/build/kubernetes/concepts/node-pools-and-autoscaling) of 2–20 nodes with labels, taints and optional autoscaling            |
| **Networking**              | Dedicated VPC per cluster, Cilium CNI, optional Traefik ingress and MetalLB — see [Networking](/products/build/kubernetes/concepts/networking)     |
| **Storage**                 | Optional dedicated storage nodes with Longhorn block volumes — see [Storage](/products/build/kubernetes/concepts/storage)                          |
| **Apps**                    | One-click App Hub (Argo CD, Grafana, PostgreSQL, WordPress, …) from the dashboard                                                                  |
| **Access**                  | Admin kubeconfig download; full `kubectl` / Helm / CI compatibility                                                                                |
| **API**                     | Full [REST API](/api-reference/kubernetes/list-clusters) — clusters, pools, kubeconfig, metrics                                                    |
| **Billing**                 | Per worker node — see [Billing model](/products/build/kubernetes/concepts/billing-model) and [Pricing](/products/build/kubernetes/details/pricing) |
| **SLA & support**           | [99.90% monthly](/products/build/kubernetes/details/sla) (endpoint path with HA) · 24/7 support                                                    |
| **Region**                  | us-east                                                                                                                                            |

## Get started

<CardGroup cols={2}>
  <Card title="Create a cluster" icon="circle-plus" href="/products/build/kubernetes/quickstart-guides/create-a-cluster">
    From zero to a running cluster in 10–20 minutes
  </Card>

  <Card title="Connect with kubectl" icon="terminal" href="/products/build/kubernetes/quickstart-guides/connect-with-kubectl">
    Download the kubeconfig and run your first commands
  </Card>

  <Card title="Scale a node pool" icon="arrow-up-right-dots" href="/products/build/kubernetes/quickstart-guides/scale-a-node-pool">
    Add capacity by hand or let the autoscaler do it
  </Card>

  <Card title="Install an app" icon="cubes" href="/products/build/kubernetes/quickstart-guides/install-an-app">
    One-click Grafana, Argo CD, PostgreSQL and more
  </Card>
</CardGroup>

## How a cluster is built

Every cluster gets its own **VPC**, a **gateway** (public IP, NAT, load-balancer entry point), one or three **control-plane nodes** (on hypervisor-local NVMe for fast etcd), your **worker pools**, and optionally dedicated **storage nodes** running Longhorn. The full picture is in [Cluster architecture](/products/build/kubernetes/concepts/cluster-architecture).