Accelerating Boolean Satisfiability (SAT) solving by common subclause elimination

Boolean Satisfiability (SAT) is an important problem in many domains. Modern SAT solvers have been widely used in important industrial applications including automated planning and verification. To solve more problems in real applications, new techniques are needed to speed up SAT solving. Inspired by the success of common subexpression elimination in programming languages and other related areas, we study the impact of common subclause elimination (CSE) on SAT solving. Intensive experiments on many SAT solvers and benchmarks with 48-h timeout show that CSE can consistently improve SAT solving. Up to 5% more SAT13 instances can be solved after CSE. LZW-based CSE shows the best overall performance, particularly in the category of application benchmarks. A potential use of this result is that one may consider the heuristic of applying CSE to boost SAT solver performance on real life applications. Because of many possible ways to improve the benefit of CSE, we hope future research can unleash the full potential of CSE in SAT solving.

[1]  Terry A. Welch,et al.  A Technique for High-Performance Data Compression , 1984, Computer.

[2]  Andrea Rendl,et al.  Effective compilation of constraint models , 2010 .

[3]  Alfred V. Aho,et al.  Compilers: Principles, Techniques, and Tools , 1986, Addison-Wesley series in computer science / World student series edition.

[4]  Stephen A. Cook,et al.  The complexity of theorem-proving procedures , 1971, STOC.

[5]  Das Amrita,et al.  Mining Association Rules between Sets of Items in Large Databases , 2013 .

[6]  Patrick Schaumont,et al.  A new algorithm for elimination of common subexpressions , 1999, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[7]  Manu Sridharan,et al.  Exploiting subformula sharing in automatic analysis of quantified formulas , 2003 .

[8]  Gilles Trombettoni,et al.  Exploiting Common Subexpressions in Numerical CSPs , 2008, CP.

[9]  Rüdiger Reischuk,et al.  On the complexity of optimal grammar-based compression , 2006, Data Compression Conference (DCC'06).

[10]  Armin Biere,et al.  Effective Preprocessing in SAT Through Variable and Clause Elimination , 2005, SAT.

[11]  Lakhdar Sais,et al.  Mining-based compression approach of propositional formulae , 2013, CIKM.

[12]  Ian P. Gent,et al.  Automatically Enhancing Constraint Model Instances during Tailoring , 2009, SARA.

[13]  Dennis P. Groth,et al.  Average-Case Performance of the Apriori Algorithm , 2004, SIAM J. Comput..

[14]  Abhi Shelat,et al.  Approximating the smallest grammar: Kolmogorov complexity in natural models , 2002, STOC '02.

[15]  Gilles Audemard,et al.  Lazy Clause Exchange Policy for Parallel SAT Solvers , 2014, SAT.

[16]  Ramakrishnan Srikant,et al.  Fast Algorithms for Mining Association Rules in Large Databases , 1994, VLDB.

[17]  James A. Storer,et al.  Data compression via textual substitution , 1982, JACM.

[18]  Martin Gebser,et al.  clasp : A Conflict-Driven Answer Set Solver , 2007, LPNMR.