Um modelo estatístico multivariado para prever o comportamento de heurísticas em verificação formal

Abstract

Functional verification is the major design-phase bottleneck for silicon productivity. Since functional verification is an NP-complete problem, it relies on a large number of heuristics with associated parameters (engines). With the advent of parallel processing, formal verification can be optimized by selecting the best n engines to run in parallel, increasing the chance of reaching verification successful termination. In this work, we present an statistical model to build engine estimators based on structural metrics and to select n engines to run in parallel. The methodology considers both engines' estimated performance and engines' correlation. Results confirmed that the methodology can be a very quick selection mechanism for parallelization of engines in order to increase the chance of running the best engines to solve the problem.