Performance analysis of constraint solvers for coverage directed test generation

Constraint random testing (CRT) is the most common verification technique used nowadays. Coverage driven constraint random test generation (CDTG) is a methodology based on CRT. In this methodology, constraints which give maximum coverage are used to randomly generate input stimuli for testing. The efficiency of CDTG depends on the technique used to generate the automated feedback path and on the constraint solver used. In this paper, a comparative study is conducted to measure the effectiveness of VCS2009.06 against other commercially available constraint solvers. Our results are obtained by solving N Queens problem and car sequencing problem using the different solvers.