On external-memory MST, SSSP and multi-way planar graph separation

  • Authors:
  • Lars Arge;Gerth Stølting Brodal;Laura Toma

  • Affiliations:
  • Department of Computer Science, Duke University;BRICS (Basic Research in Computer Science, Center of Danish National Research Foundation), University of Aarhus, Denmark;Department of Computer Science, Duke University

  • Venue:
  • Journal of Algorithms
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recently external memory graph problems have received considerable attention because massive graphs arise naturally in many applications involving massive data sets. Even though a large number of I/O-efficient graph algorithms have been developed, a number of fundamental problems still remain open.The results in this paper fall in two main classes. First we develop an improved algorithm for the problem of computing a minimum spanning tree (MST) of a general undirected graph. Second we show that on planar undirected graphs the problems of computing a multi-way graph separation and single source shortest paths (SSSP) can be reduced I/O-efficiently to planar breadth-first search (BFS). Since BFS can be trivially reduced to SSSP by assigning all edges weight one, it follows that in external memory planar BFS, SSSP, and multi-way separation are equivalent. That is, if any of these problems can be solved I/O-efficiently, then all of them can be solved I/O-efficiently in the same bound. Our planar graph results have subsequently been used to obtain I/O-efficient algorithms for all fundamental problems on planar undirected graphs.