A Distributed Parallel Algorithm for Minimum Spanning Tree Problem

TL;DR

This paper introduces a scalable distributed parallel algorithm for the minimum spanning tree problem that significantly improves performance on supercomputers by relaxing message processing constraints and employing optimization techniques.

Contribution

It presents the first parallel GHS algorithm implementation that linearly scales beyond 32 nodes, utilizing message relaxation, hashing, and compression.

Findings

Linear scalability to over 32 nodes (256 cores)

Effective message relaxation improves parallel performance

Optimization techniques enhance algorithm efficiency

Abstract

In this paper we present and evaluate a parallel algorithm for solving a minimum spanning tree (MST) problem for supercomputers with distributed memory. The algorithm relies on the relaxation of the message processing order requirement for one specific message type compared to the original GHS (Gallager, Humblet, Spira) algorithm. Our algorithm adopts hashing and message compression optimization techniques as well. To the best of our knowledge, this is the first parallel implementation of the GHS algorithm that linearly scales to more than 32 nodes (256 cores) of Infiniband cluster.