@apa512: this might be a better question for the developers of the database itself but I figure I'll ask.. say we had 100,000 documents

Maybe 1,000,000 documents

And each had a float stored 3 layers deep in the document

And we wanted to aggregate the min, max, avg and total of each of those floats across all of those documents

Is rethink even the right db for this kind of problem?