A young computer scientist and two colleagues show that searches within data structures called hash tables can be much faster than previously deemed possible.