Minimizing writes in parallel external memory search

Authors:
Nathan R. Sturtevant;Matthew J. Rutherford
Affiliations:
University of Denver, Denver, CO;University of Denver, Denver, CO
Venue:
IJCAI'13 Proceedings of the Twenty-Third international joint conference on Artificial Intelligence
Year:
2013

Citing 10
Cited 0

One jump ahead: challenging human supremacy in checkers

One jump ahead: challenging human supremacy in checkers
Ranking and unranking permutations in linear time

Information Processing Letters
A comparative analysis of parallel disk-based Methods for enumerating implicit graphs

Proceedings of the 2007 international workshop on Parallel symbolic computation
Linear-time disk-based implicit graph search

Journal of the ACM (JACM)
Using abstraction in Two-Player Games

Proceedings of the 2008 conference on ECAI 2008: 18th European Conference on Artificial Intelligence
On Computational Models for Flash Memory Devices

SEA '09 Proceedings of the 8th International Symposium on Experimental Algorithms
Compressing pattern databases

AAAI'04 Proceedings of the 19th national conference on Artifical intelligence
Best-first frontier search with delayed duplicate detection

AAAI'04 Proceedings of the 19th national conference on Artifical intelligence
Structured duplicate detection in external-memory graph search

AAAI'04 Proceedings of the 19th national conference on Artifical intelligence
Minimizing disk I/O in two-bit breadth-first search

AAAI'08 Proceedings of the 23rd national conference on Artificial intelligence - Volume 1

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recent research on external-memory search has shown that disks can be effectively used as secondary storage when performing large breadth-first searches. We introduce the Write-Minimizing Breadth-First Search (WMBFS) algorithm which is designed to minimize the number of writes performed in an external-memory BFS. WMBFS is also designed to store the results of the BFS for later use. We present the results of a BFS on a single-agent version of Chinese Checkers and the Rubik's Cube edge cubes, state spaces with about 1 trillion states each. In evaluating against a comparable approach, WMBFS reduces the I/O for the Chinese Checkers domain by over an order of magnitude. In Rubik's cube, in addition to reducing I/O, the search is also 3.5 times faster. Analysis of the results suggests the machine and state-space properties necessary for WMBFS to perform well.