Parallel mutation testing. (19th March 2012)

• Record Type: Journal Article
• Authors: Mateo, Pedro Reales; Usaola, Macario Polo
• Is Part Of: Software testing, verification & reliability. Volume 23:Number 4(2013:Oct.)
• Page Start: 315
• Page End: 350
• Publication Date: 2012-03-19
• Subjects: Computer software -- Testing -- Periodicals; Computer software -- Verification -- Periodicals; Computer software -- Reliability -- Periodicals; 005.14
• DOI: 10.1002/stvr.1471
• Languages: English
• ISSNs: 0960-0833

Abstract

<abstract abstract-type="main" id="stvr1471-abs-0001"> <title>SUMMARY</title> <p id="stvr1471-para-0001">Despite the existing techniques to reduce the costs of mutation analysis, the computational cost to apply mutation testing with large applications can be very high. One effective technique to improve the efficiency of mutation without losing effectiveness is parallel execution, where mutants and tests are executed in parallel processors, reducing the total time needed to perform mutation analysis. This paper presents a study of this technique adapted to current technologies. Five algorithms to execute mutants in parallel are analysed with three studies that use different network configurations and different number of processors with diverse characteristics. The experiments are performed with Bacterio ^P, a tool that is also presented. Unlike previous studies about parallel mutant execution, which date from the mid‐1990s, in the studies in this paper, the communication time in parallel systems no longer acts as a bottleneck. Thus, dynamic strategies, which require more communication, combined with other mutant cost reduction techniques, are the best strategies to run mutants in parallel.Copyright © 2012 John Wiley &amp; Sons, Ltd.</p> </abstract>