QuaLiKiz-jobmanager

A collection of scripts for managing many QuaLiKiz runs on Edison

Purpose

This is a collection of Python and bash scripts that can be used to manage many QuaLiKiz runs on the Edsion supercomputer. Qualikiz is a quasi-linear gyrokinetic code and can be found on GitHub.

Developing these scripts further is not planned, so this repository is more for archiving purposes. The QuaLiKiz Neural Network training set was generated using these tools.

Usage

A workflow example is given below:

1. Adjust the parameters.json to your liking. This will be used as base for all the runs.

2. Adjust the scan_parameters.csv to set up the hyperrectangle that will be scanned over.

3. Generate the skeleton for all the runs and create the job database.

    python initialize_megadb.py
   

4. Set up a folder tree to archive files in the tape storage system (hsi)

    python create_netcdf_foldertree.py
   

5. Periodically manage the jobs. For example, create a cron job that runs:

    python launch_run.py