That's excellent, thanks! I guess I didn't use the right search terms, but I was trying to search for etl, error, etc. and didn't find anything. Even better, my next question was going to be how I can fail with a meaningful message.
@GrantSmith: It wasn't about expecting improvement, my concern was making sure that the view wasn't slowing things down. As far as I know, the view is regenerated when the underlying datasets change. The 12M row dataset gets rows appended hourly, so if there's any slowness because of the view, I'm unlikely to see it.
I use DataSet View to join a 12M row dataset with 6 or 7 other smaller datasets that contain details like categories, item descriptions, etc. All are left joins, with no other changes or filters. Comparing the view to an equivalent dataset that already has the additional columns, I didn't see any appreciable difference in…