The file hashes are computed in sequence. To make it faster, it would be better to process the hashes in parallel.