Comment by rkerno

Hi, I'm curious how you deal with the potential for hash collisions across a large data set - is that a post-join check?