---
title: "Slurm"
description: "The scheduler that runs every job on the cluster, including the models."
---

> Documentation Index
> Fetch the complete documentation index at: https://docs.c4e.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Slurm

Slurm decides what runs where. Your research jobs go to the `jobs` partition, and the models run as a job of their own in the admin-only `serve` partition, which is why a job pauses them while the cluster has two Sparks. See [Slurm jobs](https://docs.c4e.ai/getting-started/slurm-jobs/).

Slurm's REST API (slurmrestd) is what lets an agent submit jobs as you through [mcp.c4e.ai](https://docs.c4e.ai/getting-started/agents/).

## Official docs

- [Slurm documentation](https://slurm.schedmd.com/documentation.html)
- [Slurm REST API](https://slurm.schedmd.com/rest.html)

Source: https://docs.c4e.ai/stack/slurm/index.mdx
