Abstract

Software Product Lines (SPL) are inherently difficult to test due to the combinatorial explosion of the number of products to consider. To reduce the number of products to test, sampling techniques such as combinatorial interaction testing have been proposed. They usually start from a feature model and apply a coverage criterion (e.g. pairwise feature interaction or dissimilarity) to generate tractable, fault-finding, lists of configurations to be tested. Prioritization can also be used to sort/generate such lists, optimizing coverage criteria or weights assigned to features. However, current sampling/prioritization techniques barely take product behavior into account. We explore how ideas of statistical testing, based on a usage model (a Markov chain), can be used to extract configurations of interest according to the likelihood of their executions. These executions are gathered in featured transition systems, compact representation of SPL behavior. We discuss possible scenarios and give a prioritization procedure illustrated on an example.
Original languageEnglish
PublisherArxiv
VolumearXiv:1310.2474
Publication statusPublished - 9 Oct 2013

Fingerprint

Sampling
Testing
Markov processes
Explosions

Keywords

  • cs.SE

Cite this

@misc{422858c5733d4706a85e1d69220b4db6,
title = "Towards Statistical Prioritization for Software Product Lines Testing",
abstract = "Software Product Lines (SPL) are inherently difficult to test due to the combinatorial explosion of the number of products to consider. To reduce the number of products to test, sampling techniques such as combinatorial interaction testing have been proposed. They usually start from a feature model and apply a coverage criterion (e.g. pairwise feature interaction or dissimilarity) to generate tractable, fault-finding, lists of configurations to be tested. Prioritization can also be used to sort/generate such lists, optimizing coverage criteria or weights assigned to features. However, current sampling/prioritization techniques barely take product behavior into account. We explore how ideas of statistical testing, based on a usage model (a Markov chain), can be used to extract configurations of interest according to the likelihood of their executions. These executions are gathered in featured transition systems, compact representation of SPL behavior. We discuss possible scenarios and give a prioritization procedure illustrated on an example.",
keywords = "cs.SE",
author = "Xavier Devroey and Maxime Cordy and Gilles Perrouin and Pierre-Yves Schobbens and Axel Legay and Patrick Heymans",
note = "Extended version published at VaMoS '14 (http://dx.doi.org/10.1145/2556624.2556635)",
year = "2013",
month = "10",
day = "9",
language = "English",
volume = "arXiv:1310.2474",
publisher = "Arxiv",
type = "Other",

}

TY - GEN

T1 - Towards Statistical Prioritization for Software Product Lines Testing

AU - Devroey, Xavier

AU - Cordy, Maxime

AU - Perrouin, Gilles

AU - Schobbens, Pierre-Yves

AU - Legay, Axel

AU - Heymans, Patrick

N1 - Extended version published at VaMoS '14 (http://dx.doi.org/10.1145/2556624.2556635)

PY - 2013/10/9

Y1 - 2013/10/9

N2 - Software Product Lines (SPL) are inherently difficult to test due to the combinatorial explosion of the number of products to consider. To reduce the number of products to test, sampling techniques such as combinatorial interaction testing have been proposed. They usually start from a feature model and apply a coverage criterion (e.g. pairwise feature interaction or dissimilarity) to generate tractable, fault-finding, lists of configurations to be tested. Prioritization can also be used to sort/generate such lists, optimizing coverage criteria or weights assigned to features. However, current sampling/prioritization techniques barely take product behavior into account. We explore how ideas of statistical testing, based on a usage model (a Markov chain), can be used to extract configurations of interest according to the likelihood of their executions. These executions are gathered in featured transition systems, compact representation of SPL behavior. We discuss possible scenarios and give a prioritization procedure illustrated on an example.

AB - Software Product Lines (SPL) are inherently difficult to test due to the combinatorial explosion of the number of products to consider. To reduce the number of products to test, sampling techniques such as combinatorial interaction testing have been proposed. They usually start from a feature model and apply a coverage criterion (e.g. pairwise feature interaction or dissimilarity) to generate tractable, fault-finding, lists of configurations to be tested. Prioritization can also be used to sort/generate such lists, optimizing coverage criteria or weights assigned to features. However, current sampling/prioritization techniques barely take product behavior into account. We explore how ideas of statistical testing, based on a usage model (a Markov chain), can be used to extract configurations of interest according to the likelihood of their executions. These executions are gathered in featured transition systems, compact representation of SPL behavior. We discuss possible scenarios and give a prioritization procedure illustrated on an example.

KW - cs.SE

M3 - Other contribution

VL - arXiv:1310.2474

PB - Arxiv

ER -