KIT | KIT-Bibliothek | Impressum

Deep Modeling through Structural Decomposition

Hinkel, Georg

Abstract:
In some applications, traditional metamodeling in two levels gets to its limits when model elements of a domain should be described as instances of other model elements. In architecture description languages, components may be instances of their component types. Although workarounds exist, these require many validation constraints and imply a cumbersome interface. To obtain more elegant metamodels that require less constraints, deep modeling seeks ways to represent non-transitive instantiation chains. However, these concepts often make existing techniques for model transformation and analysis obsolete as these languages have to be adapted. In this paper, we present an approach to realize deep modeling only through structural decomposition, which can be implemented as a non-invasive extension to meta-metamodels similar to Ecore. As a consequence, existing tools need not be adapted. We validate our concept by creating a deep modeling architecture description language and demonstrate its advantages by modeling a synthetic web application.


Zugehörige Institution(en) am KIT Forschungszentrum Informatik, Karlsruhe (FZI)
Publikationstyp Forschungsbericht
Jahr 2016
Sprache Englisch
Identifikator DOI(KIT): 10.5445/IR/1000057633
ISSN: 2190-4782
URN: urn:nbn:de:swb:90-576330
KITopen ID: 1000057633
Verlag Karlsruhe
Serie Karlsruhe Reports in Informatics ; 2016,11
Schlagworte DeepModeling; Structural Decomposition; Refinements; NMeta; Modeling Language
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft KITopen Landing Page