DocumentCode :
1123011
Title :
Parallel Game-Tree Search
Author :
Marsland, T.A. ; Popowich, Fred
Author_Institution :
Department of Computer Science, University of Alberta, Edmonton, Alta., Canada.
Issue :
4
fYear :
1985
fDate :
7/1/1985 12:00:00 AM
Firstpage :
442
Lastpage :
452
Abstract :
The design issues affecting a parallel implementation of the alpha-beta search algorithm are discussed with emphasis on a tree decomposition scheme that is intended for use on well ordered trees. In particular, the principal variation splitting method has been implemented, and experimental results are presented which show how such refinements as progressive deepening, narrow window searching, and the use of memory tables affect the performance of multiprocessor based chess playing programs. When dealing with parallel processing systems, communication delays are perhaps the greatest source of lost time. Therefore, an implementation of our tree decomposition based algorithm is presented, one that operates with a modest amount of message passing within a network of processors. Since our system has low search overhead, the principal basis for comparison is the communication overhead, which in turn is shown to have two components.
Keywords :
Algorithm design and analysis; Computer science; Councils; Decision trees; Delay effects; Delay systems; Message passing; Parallel algorithms; Parallel processing; Tree graphs; Alpha-beta search; chess programs; concurrent programming; graph and tree search strategies; message passing multiprocessors; tree decomposition;
fLanguage :
English
Journal_Title :
Pattern Analysis and Machine Intelligence, IEEE Transactions on
Publisher :
ieee
ISSN :
0162-8828
Type :
jour
DOI :
10.1109/TPAMI.1985.4767683
Filename :
4767683
Link To Document :
بازگشت