Let's say you have Drill connected to two separate databases, and you run a query where you would pull a massive amount of data from each and then do a join.
How does Drill handle this without throwing Out of Memory errors? This is assuming that the data you are requesting exceeds the amount of memory available for Drill to use.