Data Warehouse Architecture
A Data Warehouse system is the main repository of all data collected by various business divisions and departments of an enterprise. It’s architecture is made up of five main components to provide flexible reporting capabilities to an organisation.
The five main components of Data Warehouse Architecture are:
ETL = Extract – Transform – Load
The DWH server is the physical storage used to hold the data model for the enterprise data warehouse.
The main risk involves data mapping from the source systems to the data model, which ensures you have sufficient data to meet the reporting requirements.
DWH server configuration optimisation is critical in order to ensure a processing power that can handle multiple querying of data at once; while also storing, managing and securing both new and historical data. The production of reports (in various formats) on several queries, without interrupting ongoing processes, is also important
The goal of the data warehouse is to report on/publish the organization’s data assets to most effectively support decision making.
(OLAP) Online Analytical Processing is the analyses business data for the release of business insight. In order to facilitate the self-service BI requirements and complex analysis and visualisation requirements; the data in the warehouse will conform to a multi-dimensional model in the form of an OLAP cube, where data is organised into hierarchies or dimensions (with multiple levels of detail) for future/advanced business analysis and query purposes.
Using OLAP cubes, an organisation can:
For in-depth information on OLAP Cube; click HERE
Data Mining is the analysis of data sets in order to extract meaning / predict behaviours and trend analysis; and report this information in a readable format fulfilling to business user requirements.