> This page is for version 10.
> For other versions, use one of these documentation indexes:
> - 11 (default): https://docs.nvidia.com/base-command-manager/product-manuals/11/llms.txt
> - 10: https://docs.nvidia.com/base-command-manager/product-manuals/10/llms.txt

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/base-command-manager/product-manuals/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/base-command-manager/product-manuals/_mcp/server.

# Installing Slurm To An Edge Site

On some edge sites there may be a need to run a workload manager.

The Slurm workload manager can be run on an edge site if the cluster is prepared and software installed as in sections [3.1](#31-preparation) and [3.2](#32-installation).

## 3.1  Preparation

1. The edge director must be `UP` according to `cmsh` or Base View.

2. No MySQL or MariaDB installation must be present on the edge director or it will conflict with the automatic installation of `cm-mariadb`.

## 3.2  Installation

* The BCM script `cm-wlm-setup` ([section 7.3 of the *Administrator Manual*](/admin/workloadmanagement#73-installation-of-workload-managers)) is then run on the head node.

  * At the `Select installation type` screen, the edge name should be selected, and the edge director should then be specified as only server role node.

* By default there are no user home directories on the edge director or edge nodes. These must therefore be mounted or created, otherwise jobs cannot run on the edge nodes. This is true for all WLMs running on edge nodes.