Postprocessing Tool: Subvolume Data Output

[mbkuhn]

Summary

User-requested feature to output chunks of data directly from the computational mesh. Following after the work in ERF https://github.com/erf-model/ERF/blob/development/Source/IO/ERF_WriteSubvolume.cpp and taking advantage of the framework in AMR-Wind sampling (specifying many with different labels, output logic, variety of variables)

Pull request type

Please check the type of change introduced:

• Bugfix
• Feature
• Code style update (formatting, renaming)
• Refactoring (no functional changes, no api changes)
• Build related changes
• Documentation content changes
• Other (please describe):

Checklist

The following is included:

• new unit-test(s)
• new regression test(s)
• documentation for new capability

This PR was tested by running:

• the unit tests
  • on GPU
  • on CPU
• the regression tests
  • on GPU
  • on CPU

Additional background

Issue Number:

[mbkuhn]

@marchdf I haven't tested this yet, looking for feedback and improvements

Also just realized that if a user requests a non-cell-centered variable, this doesn't currently handle it properly.

[marchdf]

maybe just dx?

[marchdf]

Nice work! Yeah not sure how best to handle the non-cell centered variables. Maybe just abort until someone needs them?

Does this work if the subvolume crosses coarse-fine interfaces? If not maybe we need an abort to catch for that?

And it looks like this will only have a boxarray that matches the subvolume so the copies won't be too bad. Is that right?

[mbkuhn]

In Ann's example, there is an interpolation step because ERF's velocities are at cell faces. We could do something like that conditionally.

This doesn't traverse coarse/fine interfaces, but it uses a target resolution in the user parameters. Presumably, the user is targeting a specific level of the mesh, not multiple levels. This assumption is made in Ann's example, too.

[mbkuhn]

need a reg test and documentation for this

[mbkuhn]

not sure exactly how to do the reg test check the subvolume results, but they should be eligible inputs for fcompare

[hgopalan]

A simple reg test would be a shear layer with known velocity and computing the mean wind speed within the subvolume.

[github-actions]

This PR is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.

[github-actions]

This PR was closed because it has been stalled for 7 days with no activity.