Milan Celikovic

Mapping-Based Exchange of Models Between Meta-Modeling Tools

The exchange of models between meta-modeling tools is an important requirement. Tools often only cover a certain task in the development process. The exchange of models between different tools is necessary for covering a complete development process. Besides the aspect of cooperation, exchange of models also enables the replacement of old tools by new tools that better fit the customers needs. In order to avoid the vendor lock-in effect, model exchange allows the reuse of existing models. In this paper, we focus on the problem of model exchange and present an mapping-based approach between different meta-modeling tools. The approach is centered around a declarative mapping language with a graphical notation and a solution for connecting different meta-modeling tools. We apply our approach to exchange models between MetaEdit+ and Microsoft Visio.

A model-driven approach to data structure conceptualization

Reengineering of an existing information system can be carried out: to improve its maintainability, to migrate to a new technology, to improve quality or to prepare for functional enhancement. An important phase of a data-oriented software system reengineering is a database reengineering process and, in particular, its subprocess - a database reverse engineering process. The reverse engineering process contains two main phases: data structure extraction and data structure conceptualization. In the paper we present a blueprint of a model-driven approach to database reengineering process that is one of the results of our research project on model-driven intelligent systems for software system development, maintenance and evolution. Within that process hereinafter we focus on the data structure conceptualization process and propose a model-driven approach to data structure conceptualization. Proposed process is based on model-to-model transformations implemented by means of Atlas Transformation Language.

Extended Entity-Relationship approach in a Multi-Paradigm Information System Modeling Tool

In this paper we present a Multi-Paradigm Information System Modeling Tool (MIST) that supports Extended Entity-Relationship (EER) approach to database design. MIST components currently provide a formal specification of EER database schema specification and its transformation into the relational data model, or the class model. Also, MIST allows generation of Structured Query Language (SQL) code for database creation and procedural code for implementing database constraints. In addition, Java code that stores and processes data from the database, may be generated from the class model.

Database reverse engineering based on meta-models

Reengineering is one of the key concepts in software maintenance and evolution. It generally includes some form of reverse engineering followed by some form of forward engineering or restructuring. In the paper we focus on database reverse engineering. Model-driven software engineering promotes the idea of abstracting implementation details by focusing on: models as first class entities and automated generation of models or code from other models. In the approach meta-models are used to define the modeling languages. A database reverse engineering process can benefit of integrating meta-modeling and meta-models in the process. The plethora of models related to databases points out to the need and importance of model-to-model transformations between these models at different abstraction levels. These transformations are based on meta-models that are conformed by the source and target models of the transformations. A database reverse engineering process can be performed through a chain of model-to-model transformations based on a set of meta-models. In the paper we discuss the importance of meta-modeling in the context of database reverse engineering and present a case study illustrating an approach to database reverse engineering.

Reuse of Rules in a Mapping-Based Integration Tool

In the Internet of Things scenario, the integration of devices with business application systems requires bridging the differences in schemas of transmitted and received data. Further, different device configuration may introduce a variety in a data schema of a single device. Currently, mitigating this schema variation problem requires a manual adaptation of data transformations between the devices and business application systems. In this paper, we propose an algorithm that uses previously created transformations to automatically adjust the new ones for schema variations. The algorithm only considers isolated schema element information in order to find possible candidates in a transformation repository. Schema elements can be compared using multiple comparators, and the result is combined in a final similarity metric. Both, the algorithm and the repository are implemented as a module of AnyMap – a mapping-based integration tool. We also present a case study on which we evaluated the approach.

Faceoff: surrogate vs. natural keys

The natural and surrogate key architectures are two competing approaches to specifying keys in relational databases. We analyze the performance of these approaches with respect to the update complexity and the query data and expression complexity. Our results provide new insights into the advantages and disadvantages of both approaches.

A model-to-model transformation of a generic relational database schema into a form type data model

An important phase of a data-oriented software system reengineering is a database reengineering process and, in particular, its subprocess - a database reverse engineering process. In this paper we present one of the model-to-model transformations from a chain of transformations aimed at transformation of a generic relational database schema into a form type data model. The transformation is a step of the data structure conceptualization phase of a model-driven database reverse engineering process that is implemented in IIS*Studio development environment.