Julia stack v converting lectures from other languages

[gaelforget]

We talked a bit yesterday, with @hdrake, about whether everything needed to redo the big data lecture in native Julia was already there. Here are packages / notebooks that I am aware of and might be useful for this one or e.g. possibly lecture 5 ( @mfreilich1 , please see annotations):

• https://github.com/JuliaArrays/AxisArrays.jl
• https://github.com/JuliaArrays/LazyArrays.jl
• https://github.com/JuliaParallel/Dagger.jl , MPI.jl , etc from JuliaParallel
• NCDataSets.jl, NCTiles.jl, zarr.jl
• MeshArrays.jl (incl. C-grid -> lecture 5), https://github.com/milankl/swmone (incl. MPI -> lecture 5)
• https://github.com/milankl/ShallowWaters.jl (incl. MPI -> lecture 5)
• https://github.com/climate-machine/Oceananigans.jl (another great lecture 5 example)
• https://github.com/gaelforget/GlobalOceanNotebooks (e.g. 02_exchanges.ipynb, 03_smoothing.ipynb -> lecture 5)

Will aim to take another look / try a couple things and get back to you early next week if useful ...

ps. also e.g. the grid related plots in http://dx.doi.org/10.5194/gmd-8-3071-2015 might be representative of what climate models use (Figs 1, 2 + C1, C2 -> lecture 5)

[gaelforget]

Also (see post by @meggart in JuliaClimate/meta#1 )

• https://github.com/esa-esdl/ESDL.jl
• https://github.com/rafaqz/DimensionalData.jl

[hdrake]

I've summarized here what I think we would want as the foundation of a julia stack (basically the julia equivalent of xarray + dask), which actually may already exist in ESDL.jl but is possible still fairly unstable / poorly documented.

@meggart's PANGEO CMIP6 example is a promising start https://gist.github.com/meggart/2d544be2c1368f8774d0a21ea4633985