Implementing Framework Crosscutting Extensions with EJPs and AspectJ

Diretrizes para a Modelagem Incremental de Data Marts

Uirá KuleszaRoberta CoelhoVander AlvesAlberto Costa NetoAlessandro GarciaCarlos LucenaArndt von StaaPaulo Borba

Maria Luiza Machado CamposVânia Jesus de Araújo Soares Borges

In a previous work, we proposed a framework extension approach based on the use of a new concept, called Extension Join Points (EJPs). EJPs enable the framework systematic extension by means of variability and integration aspects. In this paper, we show how EJPs can be implemented using the mechanisms of the AspectJ language. To evaluate the usefulness of the EJPs in the framework extension process, we have used them in the development of three OO frameworks from different domains. As a result of our case studies, we present: (i) an initial categorization of different kinds of contracts between frameworks, EJPs and aspects which can be implemented in AspectJ; and (ii) a set of lessons learned when specifying the EJPs. The lack of a detailed comprehensive methodology for Data Warehouse development may be credited to fundamental differences in its requirements when compared to traditional database systems and the various approaches for the Data Warehouse architecture found in the literature. This paper presents a set of guidelines to support the incremental development of a Data Warehouse environment, created from individual Data Marts, and based on existing operational data models. These data models are integrated and transformed, deriving an initial pre-model for the Data Mart, to be refined using dimensional modeling techniques. This pre-model is the basis for the derivation and integration of each new dimensional Data Mart. Therefore, a more consistent and less empirical developing process is established.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: