Poster: Model Live Diagnostics

Title

Interactive monitoring & diagnostic analyses of live ACCESS models

About

The Model Live Diagnostics framework is a simple, easy to use and accessible Jupyter-based framework for the ACCESS modelling community to check, monitor, visualise and evaluate model behaviour and progress on currently running or ‘live’ ACCESS models on the Australian NCI supercomputer Gadi.

Read the documentation here: https://med-live-diagnostics.readthedocs.io/

Poster

Download the poster here: Tetley_ModelLiveDiagnostics.pdf (3.8 MB)

Note: this topic is part of the 2023 ACCESS Community Workshop Poster session

2 Likes

This looks great @mtetley, but not getting my vote I’m afraid. Competition is brutally strong.

@mtetley I’m very excited by this tool! Here’s the type of diagnostics I mentioned at your poster yesterday that provide a quick look at key variables to see if a coupled model run is in balance: CESM(CAM5) Atlantic PaceMaker Ensemble. This is for the CESM2 model and are global average values that are saved in the output/log file as the model runs and so the figures can get updated every timestep. These are incredibly valuable when starting a new experiment to see if everything’s going ok and the forcings are being read correctly.

There are also some timeseries in the CVDP package: Climate Variability Diagnostics Package that (I think) are run on the full model output at the end of the run.

2 Likes

Closed to release votes for this year’s posters.