Random 3-SAT and BDDs: The Plot Thickens Further
نویسندگان
چکیده
This paper contains an experimental study of the impact of the construction strategy of reduced, ordered binary decision diagrams (ROBDDs) on the average-case computational complexity of random 3-SAT, using the CUDD package. We study the variation of median running times for a large collection of random 3-SAT problems as a function of the density as well as the order (number of variables) of the instances. We used ROBDD-based pure SAT-solving algorithms, which we obtained by an aggressive application of existential quantification, augmented by several heuristic optimizations. Our main finding is that our algorithms display an “easy-hard-less-hard” pattern that is quite similar to that observed earlier for search-based solvers. When we start with low-density instances and then increase the density, we go from a region of polynomial running time, to a region of exponential running time, where the exponent first increases and then decreases as a function of the density. The locations of both transitions, from polynomial to exponential and from increasing to decreasing exponent, are algorithm dependent. In particular, the running time peak is quite independent from the crossover density of 4.26 (where the probability of satisfiability declines precipitously); it occurs at density 3.8 for one algorithm and at density 2.3 for for another, demonstrating that the correlation between the crossover density and computational hardness is algorithm dependent.
منابع مشابه
An Integrated Approach for Combining BDDs and SAT Provers
Many formal verification tools today are based on Boolean proof techniques. The two most powerful approaches in this context are Binary Decision Diagrams (BDDs) and methods based on Boolean Satisfiability (SAT). Recent studies have shown that BDDs and SAT are orthogonal, i.e. there exist problems where BDDs work well, while SAT solvers fail and vice versa. Beside this, the techniques are very d...
متن کاملAbstraction and BDDs Complement SAT-Based BMC in DiVer
ion and BDDs Complement SAT-based BMC in DiVer Aarti Gupta1, Malay Ganai, Chao Wang2, Zijiang Yang1, Pranav Ashar 1 NEC Laboratories America, Princeton, NJ, U.S.A. 2 University of Colorado, Boulder, CO, U.S.A. Abstract. Bounded Model Checking (BMC) based on Boolean Satisfiability (SAT) procedures has recently gained popularity for finding bugs in large designs. However, due to its incompletenes...
متن کاملEquivalence Checking Combining a Structural SAT-Solver, BDDs, and Simulation
This paper presents a verification technique for functional comparison of large combinational circuits using a novel combination of known approaches. The idea is based on a tight integration of a structural satisfiability (SAT) solver, BDD sweeping, and random simulation; all three working on a shared graph representation of the circuit. The BDD sweeping and SAT solver are applied in an intertw...
متن کاملRobust Boolean reasoning for equivalence checking and functional property verification
Many tasks in CAD, such as equivalence checking, property checking, logic synthesis, and false paths analysis require efficient Boolean reasoning for problems derived from circuits. Traditionally, canonical representations, e.g., BDDs, or structural SAT methods, are used to solve different problem instances. Each of these techniques offer specific strengths that make them efficient for particul...
متن کامل