Model-driven Generative Development of Measurement Software

Martin Monperrus (1) , Jean-Marc Jézéquel (2) , Benoit Baudry (2) , Joël Champeau (3) , Brigitte Hoeltzener (4)

1 Department of Computer Science
2 TRISKELL - Reliable and efficient component based software engineering
3 LISYC - Laboratoire d'Informatique des Systèmes Complexes
4 E3I2 - Extraction et Exploitation de l'Information en Environnements Incertains

Metrics offer a practical approach to evaluate non-functional properties of domain-specific models. However, it is tedious and costly to develop and maintain a measurement software for each domain specific modeling language (DSML). In this paper, we present the principles of a domain-independent, metamodel-independent and generative approach to measuring models. The approach is operationalized through a prototype that synthesizes a measurement infrastructure for a DSML. This model-driven measurement approach is model-driven from two viewpoints: 1) it measures models of a domain specific modeling language; 2) it uses models as unique and consistent metric specifications, w.r.t. a metric specification metamodel. The metric metamodel captures all the necessary concepts for model-based specifications of metrics. The specifications are used to generate a fully fledged implementation of a measurement tool. The benefit from applying the approach is evaluated by three applicative case studies. They indicate that this approach significantly reduces the domain-specific measurement software development cost with respect to code volume.

Domaines: Génie logiciel [cs.SE]

Martin Monperrus, Jean-Marc Jézéquel, Benoit Baudry, Joël Champeau, Brigitte Hoeltzener. Model-driven Generative Development of Measurement Software. Software and Systems Modeling, 2011, 10 (4), pp.537-552. ⟨10.1007/s10270-010-0165-9⟩. ⟨inria-00504670⟩