Data Architecture is about defining the structure, organization, storage, granularity, usage, quality, and maintenance of data in an enterprise. It is about understanding how the use of data relates to the business goals and practices of an enterprise in a way that produces results. A well thought architecture will provide flexibility and ease of use for the business community, providing them more accurate and timely information. It is usually identified with the various data modeling activities but also plays a vital role in other areas of data management like Data Profiling, Data Quality, Master Data Management (MDM), Data Integration, and Metadata management.
Data Architecture helps an enterprise understand the following:
• How systems and their data are aligned with related business architecture?
• What are the high level data requirements?
• How are the data requirements met conceptually/logically?
• What is the origin of the data used by the business?
• How does data flow through the various systems in the enterprise?
• Who uses the data and for what?
• How is access to data access controlled in the enterprise?
• How is the quality of data assessed and maintained?
• How is data stored, archived, retrieved?
Data Architecture consists of the following set of artifacts:
• An enterprise data model which contains information about the business entities at various levels of detail including conceptual, logical and physical views.
• A data flow model which defines the various business process that produce, consume, and transform data.
• An architectural framework that informs Data Integration, Metadata Management, Business Intelligence applications, other applications, and their interaction.