Joe Schoonover

Currently waiting on https://github.com/spack/spack/pull/44737 to merge for `feq-parse/2.2.2 `

I think this issue is related to the recent PR #7725 , at least for AMD GPUs.

Hey @cgmb - Is this ticket still relevant ?

I've started really re-working this in the https://github.com/OceanParcels/Parcels/tree/feature/uxarray_xarray_fields branch. At the moment, most of the work is being done in the [xfield.py](https://github.com/OceanParcels/Parcels/blob/feature/uxarray_xarray_fields/parcels/xfield.py) module. This module defines the `XField` and `XVectorField`...

@erikvansebille and @VeckoTheGecko - I've started summarizing some of the changes I've made (and am planning) in [this hackmd document](https://hackmd.io/@UQHn1umuRWaDx8OnpC49IQ/BkngZGi2Jl) Any feedback is welcome.

> I can imagine this changing somewhat as development progresses I agree. This has been slowly evolving as I'm working on the implementation and the stommel examples.

With us moving towards xarray/uxarray adoption, this probably won't be needed. This being said, the assumption that a `xarray.Dataset` containing all field information may not be such a bad assumption....

Documentation suggests that [`open_mfdataset`](https://docs.xarray.dev/en/stable/generated/xarray.open_mfdataset.html) (by default) will combine the datasets all the files into a single dataset using [`combine_by_coords`](https://docs.xarray.dev/en/latest/generated/xarray.combine_by_coords.html). This bit from `combine_by_coords` docs I think sums up the behavior...

[DOE's HPC Toolkit](https://hpctoolkit.org/) with [Hatchet](https://hatchet.readthedocs.io/en/latest/user_guide.html) can be quite useful for hotspot, trace, and hardware events profiling. It's open source, free, and maintained by groups in the Department of Energy's Exascale...

Hey all, with this API settling in, should we discuss concrete tasks to adapt to the new `fieldset` and `field` . I suppose something similar to @willirath 's trimmed down...