Freedom251 › Workflow automation tools
Cylc
What Cylc is
Cylc is a free workflow engine that schedules and runs computational tasks according to dependencies and timing. It supports acyclic task graphs as well as cycling workflows, including dependencies between cycles. Workflows can run across servers and platforms in network, HPC, and cloud environments, and Cylc can recover from failures and repeat workflows automatically. A web GUI, terminal interface, and command line interface let users visualise, monitor, and control runs. The dashboard can be deployed with JupyterHub for multi-user setups or alongside JupyterLab as a browser workstation. Built-in job runners include background execution, PBS or Torque, Slurm, LSF, SGE, Moab, and LoadLeveler; custom handlers can cover other runners. Multi-user permissions can grant selected users read or control access to workflows under other accounts. Cylc runs on Linux and macOS, with installation guidance recommending conda-forge. Pip installation requires Python 3.12 or later, Bash, GNU coreutils, SSH, and rsync.
Who it suits
Cylc suits weather-forecasting teams and other business or scientific research users managing scheduled computational work. It is intended for Linux or macOS environments, including distributed, HPC, and cloud setups.
What it does well
- Supports cycling and acyclic workflows
- Runs across network, HPC, and cloud systems
- Automatic recovery and workflow repetition
- Web, terminal, and command-line interfaces
- Multiple built-in job runners
What to know first
- Pip installation requires Python 3.12 or later
- Requires Bash, SSH, rsync, and GNU coreutils
- Jupyter deployments need suitable authorization policies
Plans
| Plan | Billed and limits | Price |
|---|---|---|
| Cylc | The official pages opened list no paid plans or usage limits. | Free |
As cylc.github.io lists them, read 12 Oct 2026. The maker's own page has the final word.
Verdict
Cylc provides broad scheduling, dependency, distribution, and monitoring capabilities without a paid plan. Its system requirements and security configuration need attention when deploying it for multi-user or Jupyter-based work.
Facts from Cylc's own pages
- Access controls
- For multi-user setups, the UI Server can be configured to grant specified users read or control permissions for workflows under other accounts. · cylc.github.io, 12 Oct 2026
- Conditional logic
- Yes · cylc.github.io, 24 Sept 2026
- Distributed execution
- Workflows can be distributed across servers and platforms, including network, HPC, and cloud environments. · cylc.github.io, 12 Oct 2026
- Extensibility
- Users can write custom job runner handlers when the required job runner is not among the built-in integrations. · cylc.github.io, 12 Oct 2026
- Free plan
- Yes · cylc.github.io, 24 Sept 2026
- Installation
- Cylc runs on Linux and macOS and the installation guide recommends conda-forge. · cylc.github.io, 12 Oct 2026
- Interfaces
- Cylc provides a web GUI, a terminal user interface, and a command line interface for visualising, monitoring, and controlling workflows. · cylc.github.io, 12 Oct 2026
- Job runners
- Built-in job runner handlers include background execution, PBS or Torque, Slurm, LSF, SGE, Moab, and LoadLeveler. · cylc.github.io, 12 Oct 2026
- Jupyter
- The Cylc dashboard can be deployed with JupyterHub for multi-user setups and alongside JupyterLab as a browser-based virtual workstation. · cylc.github.io, 12 Oct 2026
- Multi-step workflows
- Yes · cylc.github.io, 24 Sept 2026
- Requirements
- The pip installation requires Python 3.12 or later, Bash, GNU coreutils, SSH, and rsync. · cylc.github.io, 12 Oct 2026
- Resilience
- Cylc can automatically recover from failures and repeat workflows. · cylc.github.io, 12 Oct 2026
- Security configuration
- The documentation warns that Jupyter Server needs a suitable authorization policy in some deployments to prevent users with server access from executing arbitrary code under another account. · cylc.github.io, 12 Oct 2026
- Support
- Community support is available on the Cylc Discourse forum, and the site says commercial support is available from Altair. · cylc.github.io, 12 Oct 2026
- What it does
- Cylc is a workflow engine that automatically executes computational tasks according to schedules and dependencies. · cylc.github.io, 12 Oct 2026
- Who it is for
- Cylc was developed for weather forecasting and is described as a general-purpose tool for business and scientific research. · cylc.github.io, 12 Oct 2026
- Workflow types
- It schedules both acyclic task graphs and cycling workflows, including tasks with inter-cycle dependencies. · cylc.github.io, 12 Oct 2026
Where Cylc stands on Freedom251
- Workflow automation tools#40 of 133