作者: Lintao Zhang , Sharad Malik
DOI: 10.1007/978-3-540-24605-3_22
关键词:
摘要: We experimentally evaluate the cache performance of different SAT solvers as a case study for efficient implementation algorithms. several Boolean Constraint Propagation (BCP) mechanisms and show their respective run time performances on selected benchmark instances. From experiments we conclude that friendly data structure is key element in solvers. also empirical miss rates modern based Davis-Logemann-Loveland (DLL) algorithm with learning non-chronological backtracking. recently developed are much more structures implementations compared predecessors.