Towards platform specific energy estimation for executable domain-specific modeling languages

Thibault Béziers La Fosse (1, 2) , Massimo Tisi (1, 2) , Erwan Bousse (1, 2) , Jean-Marie Mottu (1, 2) , Gerson Sunyé (3, 1)

1 LS2N - équipe NaoMod - NaoMod - Nantes Software Modeling Group
2 IMT Atlantique - DAPI - Département Automatique, Productique et Informatique
3 LS2N - Laboratoire des Sciences du Numérique de Nantes

Energy consumption is becoming a major subject when designing, developing and running programs. Most developers code and run their applications in an energy oblivious manner, mostly because of a lack of energy-related knowledge about their system. This problem also exists in the realm of executable domain-specific modeling languages, where end-users create models conforming to a given meta-model and execute them with little knowledge about their operational semantic and related energy consumption. In this work, we propose a domain-specific language for decorating meta-models of executable languages with platform-specific energy estimation formulas. We also extend the GEMOC execution engine to dynamically perform energy estimations on any executable model conforming to the decorated meta-model. The energy estimation model defined can then be easily adapted to other models and platforms, without requiring any measurement tooling or knowledge from the end-user.

Mots clés: xDSMLs; Energy Estimation; Model-Driven Engineering

Domaines: Génie logiciel [cs.SE]; Informatique et langage [cs.CL]; Ingénierie assistée par ordinateur

Thibault Béziers La Fosse, Massimo Tisi, Erwan Bousse, Jean-Marie Mottu, Gerson Sunyé. Towards platform specific energy estimation for executable domain-specific modeling languages. MODELS-C 2019: ACM/IEEE 22nd International Conference on Model Driven Engineering Languages and Systems Companion, Sep 2019, Munich, Germany. ⟨10.1109/MODELS-C.2019.00048⟩. ⟨hal-02297501⟩