Randomized parallel list ranking for distributed memory multiprocesors

Abstract

We present a randomized parallel list ranking algorithm for distributed memory multiprocessors. A simple version requires, with high probability, (Formula presented) communication rounds (h-relations with (Formula presented) local computation. An improved version requires, with high probability, only (Formula presented) communication rounds where (Formula presented). Note that k < ln*(n) is an extremely small number. For n ≤ 1010100 and p ≥ 4, the value of k is at most 2. For a given number of processors, p, the number of communication rounds required is, for all practical purposes, independent of n. For n ≤ 1010100 and 4 ≤ p ≤ 2048, the number of communication rounds in our algorithm is bounded, with high probability, by 118. We conjecture that the actual number of communications rounds will not exceed 50.