Arrays

As discussed before, data can have more than one dimensions. We use an ArrayStore when dealing with multi-dimensional data (e.g., ZarrArrayStore and NumpyArrayStore) to ensure high performance with numerical data.

Backends

Remember, atomea is designed so that users don't have to worry about how data is stored; thus, interacting with atomea data is the exact same for every backend. Choosing a backend is made based on the specific use cases and all data can be seamlessly converted between formats.

Zarr

TODO: Introduce the who and where of zarr. Discuss when to or not to choose this array backend.

NumPy

TODO: Introduce the who and where of NPY and NPZ. Discuss when to or not to choose this array backend.