EXERCISE C notebook

[NoraLoose]

In this exercise we make a blueprint from scratch, which consists of four steps:

• Preparing the input data for a ROMS simulation (using ROMS-Tools)
• Building a C-Star Case, which includes gathering
  • ROMS source model code (cstar.base.BaseModel and cstar.base.AdditionalCode)
  • ROMS input data from step 1 (cstar.base.InputDataset)
• Compiling and running the ROMS simulation (using C-Star)
• Making a C-Star blueprint so that steps 1, 2, and 3 are reproducible

The regridding step in ROMS-Tools is currently very inefficient - in particular, when being used with use_dask=False as required on perlmutter. Even though this is a small test case, it will take the user ~40 minutes to just create the input data. That is too slow. I may have to pick up CWorthy-ocean/roms-tools#43 again, and refactor the regridding so that we re-use the same weights, even though this is technically a milestone for the beta release.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[NoraLoose]

There is now a complete notebook draft for EXERCISE C. (I'm still waiting to hear back about the account_key we should use on perlmutter.)