CWMT: A Concurrency based Acceleration Technique for Weak Mutation Testing

Mutation testing is a fault-based software testing technique, which is widely used to evaluate the adequacy of a given test suite or the fault detection effectiveness of a given software testing technique. Although mutation testing has a strong fault detection capability, the high computation cost incurred by a huge number of mutants and a long testing period prevents mutation testing from being widely adopted in practice. It can be observed that the bulk of codes in multiple mutants are repeatedly executed through an analysis of the mutation testing process. In this paper, we explore the improvement of mutation testing in terms of reducing its execution time and accordingly propose a concurrent weak mutation testing approach, which provides an optimal mechanism to removal of redundant executions by using the combination of concurrent controls and program synthesis techniques, aiming at shortening the execution time of mutation testing. An empirical study has been conducted in which 12 C programs are used as subject program were used to evaluate the effectiveness and efficiency of the proposed approach and analyze the impact factors of performance optimization. Experimental results show that the proposed approach can significantly improve the efficiency of mutation testing, with an average reduction rate of more than 90% of compilation time and 70% of execution time, respectively.