Parallel Searching for a First Solution

A parallel algorithm for conducting a search for a first solution to the problem of generating minimal perfect hash functions is presented. A message-based distributed memory computer is assumed as a model for parallel computations. A data structure, called reverse trie (r-trie), was devised to carry out the search. The algorithm was implemented on a transputer network. The experiments showed that the algorithm exhibits a consistent and almost linear speed-up. The r-trie structure proved to be highly memory efficient.