Parallel Minimax Tree Searching on GPU

  • Kamil Rocki
  • Reiji Suda
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6067)


The paper describes results of minimax tree searching algorithm implemented within CUDA platform. The problem regards move choice strategy in the game of Reversi. The parallelization scheme and performance aspects are discussed, focusing mainly on warp divergence problem and data transfer size. Moreover, a method of minimizing warp divergence and performance degradation is described. The paper contains both the results of test performed on multiple CPUs and GPUs. Additionally, it discusses αβ parallel pruning implementation.


Root Node Parallelization Scheme Minimax Algorithm Parallel Depth Thread Divergence 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Knuth, D.E., Moore, R.W.: An analysis of alpha-beta pruning. Artificial Intelligence 6, 293–326 (1975)zbMATHCrossRefMathSciNetGoogle Scholar
  2. [2]
    Manohararajah, V.: Parallel Alpha-Beta Search on Shared Memory Multiprocessors, Master Thesis, Graduate Department of Electrical and Computer Engineering, University of Toronto (2001)Google Scholar
  3. [3]
    Warren, H.S.: Hacker’s Delight. Addison-Wesley, Reading (2002)Google Scholar
  4. [4]
    Schaeffer, J.: Improved Parallel Alpha Beta Search. In: Proceedings of 1986 ACM Fall Joint Computer Conference (1986)Google Scholar
  5. [5]
    Borovska, P., Lazarova, M.: Efficiency of Parallel Minimax Algorithm for Game Tree Search. In: Proceedings of the International Conference on Computer Systems and Technologies (2007)Google Scholar
  6. [6]
    Schaeffer, J., Brockington, M.G.: The APHID Parallel αβ algorithm. In: Proceedings of the 8th IEEE Symposium on Parallel and Distributed Processing, p. 428 (1996)Google Scholar
  7. [7]
    Hewett, R., Ganesan, K.: Consistent Linear speedup in Parallel Alpha-beta Search. In: Proceedings of the ICCI 1992, Fourth International Conference on Computing and Information, pp. 237–240 (1992)Google Scholar
  8. [8]
    Hopp, H., Sanders, P.: Parallel Game Tree Search on SIMD Machines. In: Ferreira, A., Rolim, J.D.P. (eds.) IRREGULAR 1995. LNCS, vol. 980, pp. 349–361. Springer, Heidelberg (1995)Google Scholar
  9. [9]
    Sanders, P.: Efficient Emulation of MIMD behavior on SIMD Machines. In: Proceedings of the International Conference on Massively Parallel Processing Applications and Development, pp. 313–321 (1995)Google Scholar
  10. [10]
    Russell, S., Norvig, P.: Artificial Intelligence: A Modern Approach, pp. 163–171. Prentice Hall, Englewood Cliffs (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Kamil Rocki
    • 1
  • Reiji Suda
    • 1
  1. 1.JST CREST, Department of Computer Science, Graduate School of Information Science and TechnologyThe University of TokyoTokyoJapan

Personalised recommendations