How to structure inputs directories

[dougiesquire]

How do we want to structure the directories containing model input data on Gadi?

Different versions of the ACCESS-OM2 inputs include subdirectories for each component at each resolution - e.g. /g/data/ik11/inputs/access-om2/input_ff9890e2 has:

.
├── cice_01deg
├── cice_025deg
├── cice_1deg
├── common_01deg_jra55
├── common_025deg_jra55
├── common_1deg_core
├── common_1deg_jra55
├── minimal_mom_01deg
├── mom_01deg
├── mom_025deg
├── mom_1deg
├── oasis_core_to_1deg
├── yatm_01deg
├── yatm_025deg
└── yatm_1deg

Do we want to replicate this for ACCESS-OM3, or is something like the following sufficient:

.
├── 1deg # subdirectories for inputs specific to a resolution
└── share # inputs shared across configs

[micaeljtoliveira]

I would suggest to further split the inputs per component, so we can easily exclude files that are not needed for a given configuration. For example, the MOM6-CICE6 configurations don't require WW3 inputs. This could look like this:

├── 0.25deg
│   ├── cice
│   ├── mom
│   ├── share
│   └── ww3
├── 1deg
│   ├── cice
│   ├── mom
│   ├── share
│   └── ww3
├── cime
└── share

This is similar to the OM2 structure, but replacing the <comp>_<resolution> naming scheme with <resolution>/<comp> directories and renaming common to share.

[dougiesquire]

Looks sensible to me. But I'll give others an opportunity to chime in before making any changes

[aekiss]

Looks good to me. I guess the mom dirs will contain BGC inputs irrespective of whether the config uses them, but I don't see a problem with that.

[dougiesquire]

I guess the mom dirs will contain BGC inputs irrespective of whether the config uses them, but I don't see a problem with that

I think there is value in the inputs in config.yaml including only inputs that are actually used. For one, it makes the manifests more useful. We could add a bgc subdirectory within each resolution? And potentially subdirectories within that for wombat, bling... Or am I getting carried away?

[aekiss]

Yes, that would work. Would also need them for cice if we use sea ice BGC as in OM2.

[micaeljtoliveira]

I think there is value in the inputs in config.yaml including only inputs that are actually used

Yes, I fully agree.

[micaeljtoliveira]

As no one has any further comments, I think we can implement this new scheme. Nevertheless, I would wait until https://github.com/COSIMA/MOM6-CICE6-WW3/pull/4 is merged to do it.

[micaeljtoliveira]

Closing, as the proposed scheme was implemented in https://github.com/COSIMA/MOM6-CICE6/issues/10 and https://github.com/COSIMA/MOM6-CICE6-WW3/issues/8.