Welcome to kunefe’s documentation!
Indices and tables
Kunefe
Kunefe is a Python package that helps users run containerized applications (Docker) on HPC systems. It can:
convert an existing Docker image into an Apptainer image
connect to a remote host via SSH and run shell commands
generate scripts for batch jobs
submit generated job scripts to HPC batch queues (SLURM)
copy files and folders to a remote system
retrieve files and folders from a remote system
install Apptainer on a remote system
monitor the job queue (SLURM)
Requirements
In order to run your application with kunefe, you will need your application to be containerized. If your application is not containerized, you can follow this guide to learn how to do that.
The host system requirements (e.g. your laptop or desktop computer):
Remote system requirements (e.g. HPC cluster):
Note: If Apptainer is not available on the remote system, it can be installed using kunefe which will require the following packages to be installed on the remote system: curl rpm2cpio cpio
Badges
fair-software.eu recommendations |
|
---|---|
howfairis |
|
(1/5) code repository |
|
(2/5) license |
|
(3/5) community registry |
|
(4/5) citation |
|
(5/5) checklist |
Installation
To install the latest kunefe release from PyPI, run:
python -m pip install kunefe
To install the development version of kunefe from GitHub repository, do:
python -m pip install git+https://github.com/mess-nlesc/kunefe.git@main
Examples
See the examples in examples
folder.
Documentation
See https://readthedocs.org/projects/kunefe/ for the code documentation.
The project setup
The project setup is documented in project_setup.md. Feel free to remove this document (and/or the link to this document) if you don’t need it.
Contributing
If you want to contribute to the development of kunefe, have a look at the contribution guidelines.
Credits
This package was created with Cookiecutter and the NLeSC/python-template.