Triggered by a suspiciously slow join, I had a look at BATsubjoin, in particular at how it is decided which algorithm to use.

I looked at a couple of cases with a join on strings.

This is one  (pseudo syntax):

l.count ~16M 

l.T.sorted = 0

r.count = 1

In this case, a hash table on r is created (because it's smaller).

This join takes 430ms.

I forced swapping l and r, thus built the hash table on the larger bat, and then it takes 0.8ms.

Notice that though r.count = 1 is a bit of an extreme case, it does happen often in practice. And the case with a few tuples would not be very different.

The question I wanted to ask is:

is it always right to build the hash table on the small table? 

Perhaps some more heuristics can help? 

Like: if the smaller is very small, and the larger is not extremely large, then hash on the larger one?

Roberto