ForschungPublikationen
Publication Details

Details zu Publikationen

Automatic construction of parallel portfolios via algorithm configuration

verfasst von
Marius Lindauer, Holger Hoos, Kevin Leyton-Brown, Torsten Schaub
Abstract

Since 2004, increases in computational power described by Moore's law have substantially been realized in the form of additional cores rather than through faster clock speeds. To make effective use of modern hardware when solving hard computational problems, it is therefore necessary to employ parallel solution strategies. In this work, we demonstrate how effective parallel solvers for propositional satisfiability (SAT), one of the most widely studied NP-complete problems, can be produced automatically from any existing sequential, highly parametric SAT solver. Our Automatic Construction of Parallel Portfolios (ACPP) approach uses an automatic algorithm configuration procedure to identify a set of configurations that perform well when executed in parallel. Applied to two prominent SAT solvers, Lingeling and clasp, our ACPP procedure identified 8-core solvers that significantly outperformed their sequential counterparts on a diverse set of instances from the application and hard combinatorial category of the 2012 SAT Challenge. We further extended our ACPP approach to produce parallel portfolio solvers consisting of several different solvers by combining their configuration spaces. Applied to the component solvers of the 2012 SAT Challenge gold medal winning SAT Solver pfolioUZK, our ACPP procedures produced a significantly better-performing parallel SAT solver.

Externe Organisation(en)
Albert-Ludwigs-Universität Freiburg
University of British Columbia
Universität Potsdam
INRIA Institut National de Recherche en Informatique et en Automatique
Typ
Artikel
Journal
Artificial intelligence
Band
244
Seiten
272-290
Anzahl der Seiten
19
ISSN
0004-3702
Publikationsdatum
20.05.2016
Publikationsstatus
Veröffentlicht
Peer-reviewed
Ja
ASJC Scopus Sachgebiete
Sprache und Linguistik, Linguistik und Sprache, Artificial intelligence
Elektronische Version(en)
https://doi.org/10.1016/j.artint.2016.05.004 (Zugang: Offen)