Section: New Results

Formal Models for Variability

Participants : Axel Legay, Rudolf Fahrenberg, Jin Hyun Kim.

This part of the report is more concerned with task 2. It studies variability aspects in the broad scope. To simplify the study for the first year, we use the concept of software product lines. Later we shall use the results in federation of embedded systems, which is a particular class of Systems of systems.

Variability is ubiquitous in today's systems, be it in the form of configuration options or extensible architectures. By mastering variability, developers can adapt their system to changing requirements without having to develop entirely new applications. Variability is central in the context of SoS whose behaviors depend on interconnected objects. To gain information on managing variability, we have focused on Software Product Lines. Software Product Lines (SPLs) are a popular form of variability-intensive systems. They are families of similar software systems developed together to make economies of scale. SoS can be viewed as examples of product lines with interconnected obbjects. SPL engineering aims to facilitate the development of the members of a family (called products or variants) by identifying upfront their commonalities and differences. Variability in SPLs is commonly represented in terms of features, i.e., units of difference between products that appear natural to stakeholders. Each product of an SPL is therefore defined by its set of features. Hierarchies of features and dependencies between features (e.g., requires, excludes) are typically captured in a Feature Model (FM), i.e. a tree-like structure that specifies which combinations of features are valid.

[15] (C)

The model-checking problem for Software Products Lines (SPLs) is harder than for single systems: variability constitutes a new source of complexity that exacerbates the state-explosion problem. Abstraction techniques have successfully alleviated state explosion in single-system models. However, they need to be adapted to SPLs, to take into account the set of variants that produce a counterexample. In this paper, we apply CEGAR (Counterexample-Guided Abstraction Refinement) and we design new forms of abstraction specifically for SPLs. We carry out experiments to evaluate the efficiency of our new abstractions. The results show that our abstractions, combined with an appropriate refinement strategy, hold the potential to achieve large reductions in verification time, although they sometimes perform worse. We discuss in which cases a given abstraction should be used.

[18] (C)

In this work, 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 behaviour. We discuss possible scenarios and give a prioritization procedure validated on a web-based learning management software.