Incremental Model Synchronization with Precedence-Driven Triple Graph Grammars.
Technische Universität, Darmstadt
[Ph.D. Thesis], (2012)
Thesis_Marius_Lauder.pdf - Published Version
Available under Creative Commons Attribution Non-commercial No Derivatives, 2.5.
Download (6MB) | Preview
|Item Type:||Ph.D. Thesis|
|Title:||Incremental Model Synchronization with Precedence-Driven Triple Graph Grammars|
Triple Graph Grammars (TGGs) are a rule-based technique with a formal background for specifying bidirectional model transformation and, hence, can be applied to transform a given model into another and vice versa. In practice, models are either created from scratch by using a single input model, or incrementally synchronized by propagating changes between integrated models.
The outstanding property of incremental model synchronization is that in average only small portions of the whole model have to be retransformed as mostly only a subset of a model has been changed.
Hence, we have the opportunity to (i) improve efficiency of model transformations and (ii) to retain as much information as possible. Regarding information preserving capabilities, this offers the chance to qualitatively improve the results of model transformations. This is because additional model content (e.g., model elements or user specific decision during the actual transformation process), which is not covered by the model transformation itself, will be mostly retained.
In practical scenarios, unidirectional rules for incremental forward and incremental backward transformation are automatically derived from the specified TGG rules, and the overall transformation process is governed by a control algorithm. Current incremental approaches either have a runtime complexity that depends on the size of related models and not on the number of changes and their affected elements, or do not pursue formalization to give reliable predictions regarding the expected results, or impose such restrictions on the language of TGGs that the remaining expressiveness is not capable of certain real-world scenarios.
For these reasons, the aim of this thesis is to develop a novel approach to incremental model synchronization with TGGs that (i) is efficient regarding the number of changes, (ii) retains as much information as possible, (iii) complies with important formal properties, and (iv) is expressive enough for real-world scenarios.
Therefore, we introduce an incremental model synchronization algorithm for TGGs, which employs a static analysis on TGG specifications to efficiently determine the range of influence of model changes at runtime and, thus, to regard only these elements for synchronization.
Together with further improvements and critical discussions we will be able to show that this approach is a suitable means for complex model synchronization tasks.
|Place of Publication:||Darmstadt|
|Uncontrolled Keywords:||Tripelgraphgrammatik, TGG, Modellsynchronisation, Modell, Transformation, Präzedenz, Graphgrammatik, Inkrementell, eMoflon, Algorithmus, Formale Eigenschaften|
|Classification DDC:||000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik|
|Divisions:||18 Fachbereich Elektrotechnik und Informationstechnik > Institut für Datentechnik > Echtzeitsysteme|
|Date Deposited:||25 Mar 2013 14:23|
|Last Modified:||25 Mar 2013 14:23|
|Referees:||Schürr, Prof. Dr. Andy and Giese, Prof. Dr. Holger|
|Refereed:||11 February 2013|