Tools · SPBoucher
Menu

Slurm Workload Manager

by SchedMD LLC · Platform · also known as Slurm, SLURM

Open-source cluster resource manager and job scheduler for Linux clusters, allocating nodes, launching jobs and managing queues on many HPC systems

GPL-2.0-or-later licenseOpen sourceFreeSelf-hostedSelf-hostable
Visit official site Compare

Link verified Nov 2, 2026

Overview

Slurm is an open-source, fault-tolerant cluster management and job scheduling system for Linux clusters, maintained by SchedMD, which also sells support and training. Released under the GNU GPL (version 2 or later), it performs three functions: allocating exclusive or shared access to compute nodes for a period, providing a framework to start, run and monitor (usually parallel) jobs on those nodes, and arbitrating contention through a queue of pending work.

The architecture consists of a central slurmctld controller (with optional backup), a slurmd daemon on each compute node, an optional slurmdbd accounting database and a REST daemon. Users interact through sbatch, srun, squeue, sacct and scontrol. A plugin system covers scheduling policy, priority and fair share, preemption, generic resources such as GPUs, accounting, authentication, containers and network topology, and federation links several clusters.

Slurm is the dominant scheduler in academic and national HPC centers, so researchers encounter it whenever they submit jobs to a shared cluster, and Dask and Ray both provide launchers for it. It targets Linux; running it on a cluster of Macs is not a supported configuration, and installing and tuning it is a systems-administration task rather than a quick setup.

Capabilities

  • Node allocation with partitions, reservations and limits
  • Batch and interactive job launch with srun and sbatch
  • Priority, fair-share, preemption and backfill scheduling plugins
  • GPU and generic-resource scheduling
  • Accounting database and REST API
  • Multi-cluster federation

Typical workflows

  1. Submit a batch script requesting nodes, cores and GPUs
  2. Run an MPI or Dask job inside an allocation
  3. Enforce fair-share across research groups with accounting

Strengths

  • Open source with commercial support available
  • Scales from small labs to the largest supercomputers
  • Extensive plugin ecosystem

Limitations

  • Linux only; not supported on macOS nodes
  • Significant administration effort to install and tune
  • Batch-oriented; not designed for interactive services

Prerequisites

  • Linux cluster with shared authentication (MUNGE)
  • System administration access

Who it is for

  • HPC center administrators
  • Researchers submitting jobs to shared clusters

Alternatives & comparison

  • GNU Parallel

    Shell tool that runs commands in parallel over input lines on one or several machines, as a drop-in alternative to xargs and shell loops

    Open sourceOpen source
  • Kubernetes

    Open-source system for scheduling, scaling and operating containerized workloads across a cluster of machines

    Open sourceOpen source
  • Open MPI

    Open-source implementation of the MPI message-passing standard used to run tightly coupled parallel programs across the nodes of a cluster

    Open sourceOpen source
  • Ray

    Open-source Python framework for distributing tasks, actors and ML workloads across a cluster, with libraries for data processing, training, tuning, serving and reinforcement learning

    Open sourceOpen source
Compare Slurm Workload Manager with its alternatives →

Sources & verification

Primary sources consulted for this record. Claims are tied to the fields each source supports; the automated link checker re-verifies every URL on a schedule.

Confidence
High
Reviewed
Nov 1, 2026
Links
Verified · Nov 2, 2026

Spotted an error?

Broken link, outdated license, wrong category, or a tool that no longer exists? Tell us; reports are reviewed by the editor.

vLLM

vLLM project (originated at UC Berkeley Sky Computing Lab) · Local LLM inference & serving

In-depth entry

High-throughput LLM inference and serving engine with PagedAttention, continuous batching, quantization and an OpenAI-compatible server for GPU and accelerator clusters

Apache-2.0Open sourceFreeSelf-hostedNov 2, 2026

Apache Spark

Apache Software Foundation · Analytical engines & OLAP

Distributed data processing engine with DataFrame and SQL APIs in Python, Scala, Java and R, plus structured streaming and MLlib

Apache-2.0Open sourceFreeLocalSelf-hostedNov 2, 2026

Open MPI

Open MPI consortium · Distributed & parallel computing

Open-source implementation of the MPI message-passing standard used to run tightly coupled parallel programs across the nodes of a cluster

BSD-3-Clause-Open-MPIOpen sourceFreeLocalSelf-hostedBroken

Dask

Dask developers (open-source community) · Distributed & parallel computing

Python library for parallel and distributed computing that scales pandas, NumPy and custom task graphs from a laptop to a cluster

BSD-3-ClauseOpen sourceFreeLocalSelf-hostedNov 2, 2026

exo

Exo Labs · Distributed & parallel computing

Open-source software that pools several Apple Silicon Macs (and Linux hosts) into one local inference cluster, sharding large models across devices with MLX

Apache-2.0Open sourceFreeLocalSelf-hostedNov 2, 2026