Issue No. 03 - May/June (2008 vol. 10)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MITP.2008.54
Yu Lei , University of Texas, Arlington
Raghu Kacker , US National Institute of Standards and Technology
Rick Kuhn , US National Institute of Standards and Technology
Many software developers have encountered failures that occur only as the result of an interaction between two components. For example, a Web application might work correctly on Linux/Apache or Windows/IIS platforms but fail when run on a Windows XP box running an Apache server. The application error is triggered by an interaction of two components: the operating system and the Web server. Testers often use pairwise testing—all pairs of parameter values—to detect such interactions. Combinatorial testing beyond pairwise is rarely used because good algorithms for higher strength combinations, such as four-way or more, have not been available. Nevertheless, empirical evidence shows that some errors are triggered only by the interaction of three, four, or more parameters. With new algorithms and tools, developers can apply high-strength combinatorial testing capable of detecting these elusive errors.
all-pairs testing, assurance, combinatorial testing, debugging, faults, pairwise testing
Yu Lei, Raghu Kacker, Rick Kuhn, "Practical Combinatorial Testing: Beyond Pairwise", IT Professional, vol. 10, no. , pp. 19-23, May/June 2008, doi:10.1109/MITP.2008.54