The following tutorial demonstrates the parallel processing of huge (out-of-memory) raster data in the stars package. The discussed example concerns the unsupervised classification of multispectral satellite images. The used technique allows you to divide the image into smaller blocks and process them simultaneously (rather than sequentially), so you will notice an increase in performance.
