It's funny that the Cassandra blog post talks about bloom filters as the "hot new thing". They were an old idea when System R* used them to do distributed joins

It's funny that the Cassandra blog post talks about bloom filters as the "hot new thing". They were an old idea when System R* used them to do distributed joins in 1986. http://www.seas.upenn.edu/~zives/cis650/papers/r-star.pdf http://www.seas.upenn.edu/~zives/cis650/papers/r-star.pdf