We propose two topics for 2013:
Multi-view modeling
Title : A Multi-View Approach for Modeling the Aspects of a System.
Advisor : Boulanger, Frédéric, frederic.boulanger@supelec.fr
Hosting team : Supélec E3S, EA 4454
Description of the PhD thesis :
Research context
The "Modeling and Validation of Heterogeneous Systems" team at the Computer Science Department of Supélec is interested in the multi-paradigm modeling of systems in the context of Model Driven Engineering, and in the validation of the behavior of such models. It is involved in several ANR projects as well as in projects of the System@tic competitiveness cluster. It is an affiliated member of the NoE Artist2. It is deeply involved in the Gemoc initiative, whose goal is to build tools and methods for defining modeling languages in a collaborative, interoperable and composable way.
References
Benoît Combemale, Cécile Hardebolle, Christophe Jacquet, Frédéric Boulanger, Benoît Baudry
Bridging the Chasm between Executable Metamodeling and Models of Computation
Software Language Engineering, 5th International Conference SLE 2012, Revised and Selected Papers
February 2013, Krzysztof Czarnecki and Goerel Hedin editors, pages 184-203
http://wwwdi.supelec.fr/fb/publis/2012BridgingTheChasm.pdf
Marc Aiguier, Frédéric Boulanger, Bilal Kanso
A formal abstract framework for modeling and testing complex software systems
Theoretical Computer Science, vol. 455, October 2012
http://dx.doi.org/10.1016/j.tcs.2011.12.072
Frédéric Boulanger, Cécile Hardebolle, Christophe Jacquet, Dominique Marcadet
Semantic Adaptation for Models of Computations
Proceedings of the 11th International Conference on Application of Concurrency
to System Design
2011, Benoît Caillaud, Josep Carmona, Kunihiko Hiraishi editors, IEEE Computer Society, pages 153-162
http://wwwdi.supelec.fr/fb/publis/2011ACSDSemAdapt.pdf
Frédéric Boulanger, Christophe Jacquet, Cécile Hardebolle, Elyes Rouis
Modeling Heterogeneous Points of View with ModHel'X
Models in Software Engineering: Workshops and Symposia at MoDELS 2009, Denver, CO, USA
Reports and Revised Selected Papers, Sudipto Ghosh editor, 2010, Springer-Verlag, pages 310-324
http://wwwdi.supelec.fr/fb/publis/2010MultiViewModels.pdf
Short scientific description and prerequisites
Model Driven Engineering puts the focus on using models for designing and validating systems. Multi-Paradigm Modeling allows the joint use of several modeling paradigms for describing a system, which is mandatory in order to build a global model of the system, taking into account the different technical specialties of the people involved in the project, the different formalisms used at different stages of the design, and the different aspects that must be considered.
In this PhD thesis, we are studying the last point: how to model the different aspects of a system in a consistent way. Indeed, it is generally not enough for a system to behave as expected. Other aspects such as timing, memory and energy consumption or safety may be as important as the functional behavior. Although there exist some modeling techniques for taking such aspects into account, the issue of maintaining the consistency between the different models of the aspects is still open.
The objective of this PhD thesis is to propose a methodology and tools for designing and making models of different aspects of a system evolve in a consistent way during the design, each model being considered as a view on the system.
A first part will consist in a study of the current approches for multi-view modeling. Then, a generalization of these methods should lead to a multi-view modeling approach that supports any modeling paradigm. This phase should rely on the description of modeling paradigms as models of computation that has been developed in our team for several years. A last phase will apply the new approach to a case study using the ModHel'X heterogeneous modeling framework.
Prerequisites : A master in computer science or equivalent diploma.
Advantages for the PhD student
This work will allow the PhD candidate to acquire a deep knowledge of the different types of models used for designing systems, as well as the relationships between these model types. By studying different modeling paradigms, he or she will discover transverse aspects in the design of systems and gain an expertise in various domains. This global knowledge of the modeling domain will come along the mastering of tools and practices in the domain of model driven engineering. The candidate will of course participate in summer schools, workshops and conferences, and may put his work in the perspective of a funded research project.
Modular Definition of Modeling Languages
Title : Modular and Composable Definition of Modeling Languages
Advisor : Boulanger, Frédéric, frederic.boulanger@supelec.fr
Hosting team : Supélec E3S, EA 4454
Description of the PhD thesis :
Research context
The "Modeling and Validation of Heterogeneous Systems" team at the Computer Science Department of Supélec is interested in the multi-paradigm modeling of systems in the context of Model Driven Engineering, and in the validation of the behavior of such models. It is involved in several ANR projects as well as in projects of the System@tic competitiveness cluster. It is an affiliated member of the NoE Artist2. It is deeply involved in the Gemoc initiative, whose goal is to build tools and methods for defining modeling languages in a collaborative, interoperable and composable way.
References
Benoît Combemale, Cécile Hardebolle, Christophe Jacquet, Frédéric Boulanger, Benoît Baudry
Bridging the Chasm between Executable Metamodeling and Models of Computation
Software Language Engineering, 5th International Conference SLE 2012, Revised and Selected Papers
February 2013, Krzysztof Czarnecki and Goerel Hedin editors, pages 184-203
http://wwwdi.supelec.fr/fb/publis/2012BridgingTheChasm.pdf
Marc Aiguier, Frédéric Boulanger, Bilal Kanso
A formal abstract framework for modeling and testing complex software systems
Theoretical Computer Science, vol. 455, October 2012
http://dx.doi.org/10.1016/j.tcs.2011.12.072
Frédéric Boulanger, Cécile Hardebolle, Christophe Jacquet, Dominique Marcadet
Semantic Adaptation for Models of Computations
Proceedings of the 11th International Conference on Application of Concurrency
to System Design
2011, Benoît Caillaud, Josep Carmona, Kunihiko Hiraishi editors, IEEE Computer Society, pages 153-162
http://wwwdi.supelec.fr/fb/publis/2011ACSDSemAdapt.pdf
Frédéric Boulanger, Christophe Jacquet, Cécile Hardebolle, Elyes Rouis
Modeling Heterogeneous Points of View with ModHel'X
Models in Software Engineering: Workshops and Symposia at MoDELS 2009, Denver, CO, USA
Reports and Revised Selected Papers, Sudipto Ghosh editor, 2010, Springer-Verlag, pages 310-324
http://wwwdi.supelec.fr/fb/publis/2010MultiViewModels.pdf
Short scientific description and prerequisites
Model Driven Engineering puts the focus on using models for designing and validating systems. Multi-Paradigm Modeling allows the joint use of several modeling paradigms for describing a system, which is mandatory in order to build a global model of the system, taking into account the different technical specialties of the people involved in the project, the different formalisms used at different stages of the design, and the different aspects that must be considered.
In this PhD thesis, we are studying the definition of modeling languages, which has two main aspects: syntax and semantics. The syntactic aspects of a language are well supported by numerous tools. On the contrary, there are very few standard tools for defining the semantics of a language, and this is still a difficult task. In order to ease the design of the various aspects of a modeling language (data processing, control structures, handling of time), our goal is to setup a framework that allows the modular definition of these aspects and their reuse in different languages.
A first part of the thesis will consist in the study of the existing approaches for the definition of modeling languages, both for syntax and semantics. The next step will be to identify the concepts around which the different aspects of a modeling language are connected, in order to define them in a modular way. Then, we will extend the composition mechanisms of the different aspects of a language to target the compositions of modeling languages. A last phase will validate the new approach by applying it to at least two DSLs in the context of the Gemoc initiative.
Prerequisites : A master in computer science or equivalent diploma.
Advantages for the PhD student
This work will allow the PhD candidate to acquire a deep knowledge of the different types of models used for designing systems, as well as the relationships between these model types. By studying different modeling paradigms, he or she will discover transverse aspects in the design of systems and gain an expertise in various domains. This global knowledge of the modeling domain will come along the mastering of tools and practices in the domain of model driven engineering. The candidate will of course participate in summer schools, workshops and conferences, and may put his work in the perspective of a funded research project.