A data architecture, in simple terms, is a framework for the IT infrastructure to be able to support the data strategy. It refers to the models, rules through which the data is collected, arranged, stored, transported, and utilized in an organization. Data capabilities include metadata management, master/reference data management, data integration, data privacy, analytics, and architecture. Data architecture framework reflects how each capability fits into the overall data management.
This video explains the generic analytical data architecture framework & how the data flows from data producers and transforms into meaningful insight.
Data architecture has many layers based on data management capabilities.
Data flow starts with the operational data source, also called transactional data. It is followed by data integration which is about the data movement between source and target systems. There could be multiple layers. There are two static components in the architecture – Data Lake and Data Mart. Data Lake is where the data is integrated, and Data Marts are the subject-oriented data assets. Finally, an analytics layer provides the reporting and analysis capabilities to the business users.
The shared services ensure the effectiveness and consistency of data architecture.