Pairwise feature-interaction testing for SPLs: potentials and limitations

  • Authors:
  • Sebastian Oster;Marius Zink;Malte Lochau;Mark Grechanik

  • Affiliations:
  • TU Darmstadt, Darmstadt, Germany;TU Darmstadt, Darmstadt, Germany;TU Braunschweig, Braunschweig, Germany;Accenture Labs & UIC, Chicago, IL

  • Venue:
  • Proceedings of the 15th International Software Product Line Conference, Volume 2
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

A fundamental problem of testing Software Product Lines (SPLs) is that variability enables the production of a large number of instances and it is difficult to construct and run test cases even for SPLs with a small number of variable features. Interacting features is a foundation of a fault model for SPLs, where faults are likely to be revealed at execution points where features exchange information with other features or influence one another. Therefore, a test adequacy criterion is to cover as many interactions among different features as possible, thus increasing the probability of finding bugs. Our approach combines a combinatorial designs algorithm for pairwise feature generation with model-based testing to reduce the size of the SPL required for comprehensive coverage of interacting features. We implemented our approach and applied it to an SPL from the automotive domain provided by one of our industrial partners. The results suggest that with our approach higher coverage of feature interactions is achieved at a fraction of cost when compared with a baseline approach of testing all feature interactions.