Parallel Retrograde Analysis on a Distributed System

Parallel Retrograde Analysis on a Distributed System (1995)

Cached

Download Links

[ftp.cs.vu.nl]

Other Repositories/Bibliography

DBLP

by Henri Bal , Victor Allis

Venue:	In Supercomputing '95
Citations:	17 - 10 self

Versions

Version History

Metadata Version 2

Datum	Value	Source
TITLE	Parallel Retrograde Analysis on a Distributed System	SVM HeaderParse 0.1
AUTHOR NAME	Henri Bal	SVM HeaderParse 0.1
AUTHOR AFFIL	Dept. of Mathematics and Computer Science; Vrije Universiteit	SVM HeaderParse 0.2
AUTHOR ADDR	De Boelelaan 1081a; 1081 HV Amsterdam, The Netherlands	SVM HeaderParse 0.1
AUTHOR NAME	Victor Allis	SVM HeaderParse 0.1
AUTHOR AFFIL	Dept. of Mathematics and Computer Science; Vrije Universiteit	SVM HeaderParse 0.2
AUTHOR ADDR	De Boelelaan 1081a; 1081 HV Amsterdam, The Netherlands	SVM HeaderParse 0.1
ABSTRACT	Retrograde Analysis (ra) is an AI search technique used to compute endgame databases, which contain optimal solutions for part of the search space of a game. ra has been applied successfully to several games, but its usefulness is restricted by the huge amount of cpu time and internal memory it requires. We present a parallel distributed algorithm for ra that addresses these problems. ra is hard to parallelize efficiently, because the communication overhead potentially is enormous. We show that the overhead can be reduced drastically using message combining. We implemented the algorithm on an Ethernet-based distributed system. For one example game (awari), we have computed a large database in 50 minutes on 64 processors, whereas one machine took 40 hours (a speedup of 48). An even larger database (computed in half a day) would have required 400 MByte of internal memory on a uniprocessor and would compute for weeks.	SVM HeaderParse 0.1
YEAR	1995	INFERENCE
VENUE	In Supercomputing '95	INFERENCE
VENUE TYPE	CONFERENCE	INFERENCE
CITATIONS	14 found	ParsCit 1.0