| Resumo: | Abstract The emergence of new areas, such as the internet of things, which require access to the latest data for data analytics and decision-making environments, created constraints for the execution of analytical queries on traditional data warehouse architectures. In addition, the increase of semi-structure and unstructured data led to the creation of new databases to deal with these types of data, namely, NoSQL databases. This led to the information being stored in several different systems, each with more suitable characteristics for different use cases, which created difficulties in accessing data that are now spread across various systems with different models and characteristics. In this work, a system capable of performing analytical queries in real time on distributed and heterogeneous data sources is proposed: EasyBDI. The system is capable of integrating data logically, without materializing data, creating an overview of the data, thus offering an abstraction over the distribution and heterogeneity of data sources. Queries are executed interactively on data sources, which means that the most recent data will always be used in queries. This system presents a user interface that helps in the configuration of data sources, and automatically proposes a global schema that presents a generic and simplified view of the data, which can be modified by the user. The system allows the creation of multiple star schemas from the global schema. Finally, analytical queries are also made through a user interface that uses drag-and-drop elements. EasyBDI is able to solve recent problems by using recent solutions, hiding the details of several data sources, at the same time that allows users with less knowledge of databases to also be able to perform real-time analytical queries over distributed and heterogeneous data sources.
|
|---|