Superlinear speedup in parallel state-space search

  • V. Nageshwara Rao
  • Vipin Kumar
Session 5 Parallel Algorithms
Part of the Lecture Notes in Computer Science book series (LNCS, volume 338)


When N processors perform depth-first search on disjoint parts of a state space tree to find a solution, the speedup can be superlinear (i.e., > N) or sublinear (i.e., <N) depending upon when a solution is first encountered in the space by one of the processors. It may appear that on the average, the speedup would be either linear or sublinear. Using an analytical model, we show that if the search space has more than one solution and if these solutions are randomly distributed in a relatively small region of the search space, then the average speedup in parallel depth-first search can be superlinear. If all the solutions (one or more) are uniformly distributed over the whole search space, then the average speedup is linear. This model is validated by our experiments on synthetic state-space trees and the 15-puzzle problem. The same model predicts average superlinear speedup in parallel best-first branch-and-bound algorithms on suitable problems.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    M. Imai, Y. Yoshida, and T. Fukumura. A parallel searching scheme for multiprocessor systems and its application to combinatorial problems. In IJCAI, pages 416–418, 1979.Google Scholar
  2. [2]
    Virendra K. Janakiram, Dharma P. Agrawal, and Ram Mehrotra. Randomized parallel algorithms for prolog programs and backtracking applications. In Proceedings of International conference on Parallel Processing, pages 278–281, 1987.Google Scholar
  3. [3]
    Laveen Kanal and Vipin Kumar (editors). Search in Artificial Intelligence. Springer-Verlag, New York, 1988 (in press).Google Scholar
  4. [4]
    R.E. Korf. Depth-first iterative-deepening: an optimal admissible tree search. Artificial Intelligence, 27:97–109, 1985. Also a chapter in 'search and Artificial Intelligence', Vipin Kumar and Laveen Kanal Eds, Springer-Verlag,1987(to appear).Google Scholar
  5. [5]
    Richard Korf. Personal communication. In the Univ. of California at Los Angeles, 1988.Google Scholar
  6. [6]
    W. Kornfeld. The use of parallelism to implement a heuristic search. In IJCAI, pages 575–580, 1981.Google Scholar
  7. [7]
    Vipin Kumar. Depth-first search. In Stuart C. Shapiro, editor, Encyclopaedia of Artificial Intelligence: Vol 2, pages 1004–1005, John Wiley and Sons, Inc., New York, 1987.Google Scholar
  8. [8]
    Vipin Kumar and Laveen N. Kanal. Parallel branch-and-bound formulations for and/or tree search. IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-6, November 84.Google Scholar
  9. [9]
    Vipin Kumar, K. Ramesh, and V. Nageshwara Rao. Parallel heuristic search of state-space graphs: a summary of results. In Proceedings of the 1988 National Conference on Artificial Intelligence, August 1988. Also AI Lab Tech. Report 88-70, University of Texas at Austin, March 88.Google Scholar
  10. [10]
    Vipin Kumar and V. Nageshwara Rao. Parallel depth-first search, part II: analysis. International Journal of Parallel Programming, 16 (6), 1987.Google Scholar
  11. [11]
    Kouichi Kumon, Hideo Masuzawa, and Akihiro Itashaki. Kabu-wake: a new parallel inference method and its evaluation. In Digest of papers — IEEE Computer Society International Conference, pages 168–172, 1986.Google Scholar
  12. [12]
    T. H. Lai and Sartaj Sahni. Anomalies in parallel branch and bound algorithms. Communications of the ACM, 594–602, 1984.Google Scholar
  13. [13]
    Guo-Jie Li and Benjamin W. Wah. Computational Efficiency of Parallel Approximate Branch-and-Bound Algorithms. Technical Report TR-84-6, Purdue University, School of Electrical Engineering, FEBRUARY 84.Google Scholar
  14. [14]
    Guo-Jie Li and Benjamin W. Wah. Coping with anomalies in parallel branch-and-bound algorithms. IEEE Trans on Computers, C-35, June 1986.Google Scholar
  15. [15]
    R. Mehrotra and E. Gehringer. Superlinear speedup through randomized algorithms. In Proceedings of International conference on Parallel Processing, pages 291–300, 1985.Google Scholar
  16. [16]
    Joseph Mohan. Experience with two parallel programs solving the traveling salesman problem. In Proceedings of International conference on Parallel Processing, pages 191–193, 1983.Google Scholar
  17. [17]
    B. Monien, E. Spekenmeyer, and O. Vornberger. Superlinear Speedup for Parallel Backtracking. Technical Report 30, Univ. of Paderborn, FRG, 1986.Google Scholar
  18. [18]
    V. Nageshwara Rao, V. Kumar, and K. Ramesh. A parallel implementation of iterative-deepening-a*. In AAAI, pages 878–882, 1987. Also AI Lab TR 87-46, University of Texas at Austin, January 87.Google Scholar
  19. [19]
    V. Nageshwara Rao and Vipin Kumar. Parallel depth-first search, part I: implementation. International Journal of Parallel Programming, 16 (6), 1987.Google Scholar
  20. [20]
    D.S. Nau, V. Kumar, and L. Kanal. General branch-and-bound and its relation to a* and ao*. Artificial Intelligence, 23, 1984.Google Scholar
  21. [21]
    Nils J. Nilsson. Principles of Artificial Intelligence. Tioga Press, 1980.Google Scholar
  22. [22]
    Judea Pearl. Heuristics — Intelligent Search Strategies for Computer Problem Solving. Addison-Wesley, Reading, MA, 1984.Google Scholar
  23. [23]
    Michael J. Quinn and Narsingh Deo. An upper bound for the speedup of parallel branch-and-bound algorithms. BIT, 6,No 1, March 1986.Google Scholar
  24. [24]
    H. Stone and P. Sipala. The average complexity of depth-first search with backtracking and cutoff. IBM Journal of Research and Development, May 1986.Google Scholar
  25. [25]
    Peter Tinker. Performance and pragmaticsof an OR-parallel logic programming system. International Journal of Parallel Programming, ?, 1988.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • V. Nageshwara Rao
    • 1
  • Vipin Kumar
    • 1
  1. 1.Department of Computer SciencesUniversity of Texas at AustinAustin

Personalised recommendations