Using Containers

Docker provides the easiest way of installing. We recommend this for users, who do not have access to a Linux system or Window Subsystem for Linux (WSL) or need to transport their code onto supercomputers. Of course, the Docker service must be available on the system, where computations with 13CFLUX are supposed to be executed.

The provided Docker image acts as a blueprint of a system with a ready installation and is instantiated as a container. Containers are similar to a running virtual machine, but leverage the local hardware more efficiently.

To get the Docker image with pre-installed 13CFLUX, simply run

docker pull jugit-registry.fz-juelich.de/ibg-1/modsim/fluxomics/13cflux

in a terminal or by using the URL in the Docker client GUI.

A blueprint to execute jobs with 13CFLUX is given by

docker run -v /LOCAL/PATH:/task --rm --name x3cflux-container \
            jugit-registry.fz-juelich.de/ibg-1/modsim/fluxomics/13cflux \
            /opt/python/cp313-cp313/bin/python /task/SOME_SCRIPT.py

which instantiates a container from an image, maps the external path /LOCAL/PATH to the internal path /task and executes the script SOME_SCRIPT.py that was mounted in the container. Within the script, 13CFLUX is executed, results are stored somewhere /task and are thereby also available on the local filesystem after the container has stopped. Omit the --rm option, if you want to use the same container multiple times.