Parallel minimax tree searching on GPU

Authors:
Kamil Rocki;Reiji Suda
Affiliations:
JST, CREST, Department of Computer Science, Graduate School of Information Science and Technology, The University of Tokyo, Tokyo, Japan;JST, CREST, Department of Computer Science, Graduate School of Information Science and Technology, The University of Tokyo, Tokyo, Japan
Venue:
PPAM'09 Proceedings of the 8th international conference on Parallel processing and applied mathematics: Part I
Year:
2009

Citing 7
Cited 0

Improved parallel alpha-beta search

ACM '86 Proceedings of 1986 ACM Fall joint computer conference
Hacker's Delight

Hacker's Delight
Consistent Linear Speedup in Parallel Alpha-Beta Search

ICCI '92 Proceedings of the Fourth International Conference on Computing and Information: Computing and Information
Parallel Game Tree Search on SIMD Machines

IRREGULAR '95 Proceedings of the Second International Workshop on Parallel Algorithms for Irregularly Structured Problems
Artificial Intelligence: A Modern Approach

Artificial Intelligence: A Modern Approach
The APHID Parallel alpha-beta Search Algorithm

SPDP '96 Proceedings of the 8th IEEE Symposium on Parallel and Distributed Processing (SPDP '96)
Efficiency of parallel minimax algorithm for game tree search

CompSysTech '07 Proceedings of the 2007 international conference on Computer systems and technologies

Quantified Score

Hi-index	0.00

Visualization

Abstract

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.