I remember reading this and getting excited. Still today we haven't quite managed to paralellize all the things, even across cpu cores much less nodes.