Partitioning Strategies for Distributed SMT Solving
share
For many users of Satisfiability Modulo Theories (SMT) solvers, the solver's performance is the main bottleneck in their application. One promising approach for improving performance is to leverage the increasing availability of parallel and cloud computing. However, despite many efforts, the best parallel approach to date consists of running a portfolio of solvers, meaning that performance is still limited by the best possible sequential performance. In this paper, we revisit divide-and-conquer approaches to parallel SMT, in which a challenging problem is partitioned into several subproblems. We introduce several new partitioning strategies and evaluate their performance, both alone as well as within portfolios, on a large set of difficult SMT benchmarks. We show that hybrid portfolios that include our new strategies can significantly outperform traditional portfolios for parallel SMT.
READ FULL TEXT