In this paper, we show how model transformations can be used to implement data distribution features in the software design process of a component. This approach is based on a single metamodel that defines data distribution abstractions and on the design of alternatives that are used to implement each data distribution variant. A model transformation is associated with the metamodel and the component metamodel we consider as the target. We show that this approach facilitates the derivation of different implementation strategies from the model of a component. We illustrate our approach with the example of distributed communication component software that implements one centralized and two peer-to-peer variants and we demonstrate the reusability of the transformation. © 2008 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Kaboré, E., & Beugnard, A. (2008). Implementing a data distribution variant with a metamodel, some models and a transformation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5053 LNCS, pp. 224–237). https://doi.org/10.1007/978-3-540-68642-2_18
Mendeley helps you to discover research relevant for your work.