I love that drake can async multiple tasks at once, but it would be cool if I can leverage that further and take a single file of data where each 'row' is encapsulated, split that out into batches and then have Drake work on each file in parallel. Then in a later step, take those worked on batch files output and merge them all back together.
Is this currently possible in some way, or will it be in the future? Or is it something outside the scope of Drake?
Thanks.