In our ongoing series on geospatial raster data formats, Julia Signell and I have been exploring the finer points of array data storage. Throughout our research, we’ve found that chunking – breaking a large dataset down into smaller pieces for individual storage and retrieval – is universally relevant regardless of data format. Chunking, as we’ve seen, is an absolutely necessary strategy for making large datasets usable, but in the cloud era, it has become something tyrannical, making...
