My workflow generate a dataset as `xr.Dataset with dims (6, 36, 2, 13, 699, 1920) in float32. I can process and write output array chunk by chunk if already zarr is existing with ds.to_zarr('data.zarr', region=region). Does anyone have an idea how to initialize the zarr file which is larger than available memory? I am using following version of libraries. zarr-python: '2.18.4' xarray: '2025.1.2' Continue reading...
