Input archive

The Input archive must be a zip file. The number of jobs created depends on the data contained in the Input archive. Each created job is executed in parallel and has the resources you requested.

Solving Flux project(s)

In order to solve Flux project(s), add the .FLU directory to the archive as below:

One job will be created per Flux project and will solve every scenario defined. Hence, the projects must contain at least one scenario.

Post-processing

To compute post-processing work on Flux projects, add one python named postpro.py, as below:

The created jobs will load the projects and execute the python script. Thus, the same python script is executed for all the projects.

Custom jobs

To compute a custom job for a particular project you can add a python script with the same name. For example, to create two jobs doing different computations, you can associate one python to both Flux projects as below:

This case extends the two previous cases, i.e. if a project does not have an associated python, the postpro.py is used if it exists, otherwise, the job will solve all its scenarios. So, the archive below are still valid:

Arch3a.zip will create two jobs: one executing MyProject_1.py and the other solving MyProject_2.FLU, whereas Arch3b.zip loads MyProject_2.FLU and executes postpro.py.

Python execution jobs

For more advanced users, one can provide an archive full of python scripts and other files. One job will be created for each main*.py (where * is a number), for example:

will create three jobs, each executing a main*.py. The other files should not contain any Flux projects, or one of the three previous cases will be considered instead.