Payu

Current release: 1.3.2

Payu is a workflow management tool for running numerical models in supercomputing environments.

Payu was designed to allow users to start running climate models immediately, without having to re-learn the nuances of countless runscripts across countless models. Running a model like the MOM ocean model should only require a few commands. First, create a new experiment directory:

mkdir new_expt; cd new_expt

Next, Create experiment or clone an existing released configuration and customise it as needed. For example, to clone MOM5 test configuration:

payu clone -b control -B master https://github.com/payu-org/bowl1.git bowl1
cd bowl1

Then generate the executable file by following the README instructions.

To run the model, simply execute:

payu run

Currently, payu is very highly customised for users of NCI computing environments, with a very strong dependence on environment modules, the PBS scheduler, and an MPI runtime environment. Using payu on other machines will require, at a minimum, the installation of these services, as well as a potentially significant modification of the codebase.

User Guide

Contents:

• Installation
  • NCI Users
  • General Use
• Usage
  • Overview
  • Setting up the laboratory
  • Running your experiment
  • Monitoring payu jobs
  • Cleaning up
  • Postprocessing
  • Metadata and Related Experiments
  • Common flags
  • Getting support
• Configuring your experiment
  • Configuration Settings
  • Miscellaneous
• Manifests
  • Introduction
  • Manifest contents
  • Experiment tracking
• Design Notes
  • Model and Experiment Layout
  • Experiment Steps
  • Post-processing PBS Jobs
  • Style Guide

Support and Development

Support is coordinated through issues, discussions and pull requests on the GitHub repository https://github.com/payu-org/payu