Deterministic Parallel DPLL

Youssef Hamadi (1, 2, 3) , Said Jabbour (4) , Cédric Piette (4) , Lakhdar Saïs (4)

1 MSR - INRIA - Microsoft Research - Inria Joint Centre
2 LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]
3 Microsoft - Microsoft Research [Cambridge]
4 CRIL - Centre de Recherche en Informatique de Lens

Current parallel SAT solvers su er from a non-deterministic behavior. This is the consequence of their architectures which rely on weak synchronizing in an attempt to maximize performance. This behavior is a clear downside for practitioners, who are used to both runtime and solution reproducibility. In this paper, we propose the rst Deterministic Parallel DPLL engine. Our experimental results clearly show that our approach preserves the performance of the parallel portfolio approach while ensuring full reproducibility of the results.

Domaines: Intelligence artificielle [cs.AI]

Youssef Hamadi, Said Jabbour, Cédric Piette, Lakhdar Saïs. Deterministic Parallel DPLL. Journal on Satisfiability, Boolean Modeling and Computation, 2011, 7 (4), pp.127-132. ⟨hal-00868187⟩