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

# Execution backends

> Connect Node, Kubernetes, or Slurm Fleets to a CPU-only control plane.

The control plane handles authentication, scheduling, metadata, and the console. It needs no GPU or Docker socket. Compute resources are registered as **Fleets**, independently of where the control plane runs.

| Backend | Register | Execution |
| - | - | - |
| `node` | Enroll compute machines in a Node Fleet | Containers on one to eight machines per job |
| `kubernetes` | Register an existing Kubernetes cluster | KubeRay workloads managed by that cluster |
| `slurm` | Register an existing Slurm cluster | Allocations managed by Slurm |

A single compute machine also uses a Node Fleet. Install the Node daemon on that machine and enroll it using a Join Token. There is no control-plane `local` executor.

## Serving

Node and Kubernetes Fleets support both temporary Playground sessions and Model Deployments. Node reserves host ports and returns the selected machine's workload address; ensure the control plane can reach the serving port. Kubernetes uses its API proxy. Slurm does not provide interactive serving.

## Setup

> **[Node](https://docs.tuneplane.ai/en/ops/executor-node.md)**
>
> Enroll one or more compute machines.

> **[Kubernetes](https://docs.tuneplane.ai/en/ops/executor-kubernetes.md)**
>
> Connect an existing Kubernetes cluster.

> **[Slurm](https://docs.tuneplane.ai/en/ops/executor-slurm.md)**
>
> Connect an HPC cluster.

Deploy the control plane with [Docker Compose](https://docs.tuneplane.ai/en/ops/install-compose.md) or [Kubernetes](https://docs.tuneplane.ai/en/ops/install-kubernetes.md), then [register Fleets](https://docs.tuneplane.ai/en/ops/fleets.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.