The Data Vault modeling method is designed to support multi-source enterprise data warehouses, which require rapid adaptation to new and changing business requirements. But the implementation of a Data Vault can be a complex and time-consuming process. Enabling data warehouse teams to achieve agility in their projects and deliver flexible, affordable and robust solutions in minimum amount of time, biGENiUS now provides comprehensive automation for Data Vault.
Data Vault modeling, originally invented by Dan Linstedt, aims to address challenges which modern data warehousing faces in today's volatile business environment. Its primary focus is to provide flexible data warehouse solutions, that have 'the ability to adapt quickly, model the business accurately, and scale with the business needs – converging IT and the business to meet the goals of the corporation' Furthermore, the data vault concept deals with other aspects of modern data warehousing like agile development, information integration, load performance, data traceability and auditability.
Data Vault is typically used for data modeling of the core layer of a data warehouse or building an enterprise data warehouse (EDW) with many different source systems. The basic concept is to define information in such a way, so that new data can be easily integrated, accessed and historized over time. A main principle of data vault is to define business entities into following core components:
- Hubs for the business keys of business entities
- Links for the relationships between business entities
- Satellites for the historized attributes of an entity
With these three types of tables – Hubs, Links and Satellites – comprehensive and extensible data models can be built. A Link may refer to two or more Hubs, and multiple Satellites may refer to one Hub. Each Hub, Link and Satellite is stored in a separate database table. As a result, data vault models usually contain a high number of tables, for each of which a corresponding ETL process is required. Accordingly, a high number of load processes need to be implemented, as well. But the number of distinct design patterns remains rather small and the loaders for all the tables of a particular table type are structured in the same way.
Generate the Data Vault
The data vault methodology combines elegantly with data warehouse automation (DWA). Using DWA tools helps you avoid boring and repetitive manual development tasks, so that your data warehousing team can focus on the needs of business decision-makers. In case of data vault, using automation is not only recommended, but rather crucial for the success - especially when the team members are not experienced with this rather detail-oriented and complex methodology.
biGENiUS is designed to provide data warehouse automation and lifecycle management support for various data modelling methods, including Data Vault. The tool uses a matured metadata-driven approach to ensure consistency and flexibility across the entire lifecycle of a data warehouse. Once the metadata is captured in the biGENiUS application, all required data vault-specific objects, tables and load processes are easily generated from scratch. As a result, implementation times and error rates decrease, while the quality of the code and productivity of your development team increase.
More than automation of development tasks
Drawn on the practical experience from many and various BI projects, biGENiUS incorporates best practices for the DWH architecture. These built-in standards can be adopted as they are, but also easily customized to your specific project needs. For instance, in case you need to define new naming conventions, or add new default columns or create even additional DWH layers such as a separate Raw and Business Data Vault. biGENiUS offers you a broad, but at the same time user-friendly configurability.
What's more, biGENiUS provides more than just generation of DWH objects. The tool will also ensure the long-term ease of maintenance of the generated data warehouse solution – through improved consistency, better and always up-to-date documentation, simplified testing, version control, impact analysis, automated implementation of changes and standardized deployment methodology.