Kimball is a data warehouse design methodology developed by Ralph Kimball. The Kimball methodology focuses on designing data warehouses that can be effectively used by business users to make decisions based on reliable and consolidated data.
The Kimball methodology emphasizes designing a data warehouse as a collection of dimensional models, as opposed to relational models used in operational systems. Dimensional models are designed to provide quick and easy access to specific data, such as customer behavior, sales performance or financial results. Often the terms “facts” and “dimensions” are used here:
-
Facts: a fact is a measurable event that occurs in a business process. For example, it can be the sale of a product, a customer transaction, or the delivery of an order. Facts are measured in numeric values and are often associated with one or more dimensions to specify the context of the event.
-
Dimensions: A dimension is a descriptive category that gives context to a fact. Dimensions can include several categories, such as time, location, product, customer, etc. Dimensions contain the characteristics or attributes of these categories. For example, the dimension “time” may include the date, quarter, month and day of the week.
A key aspect of the Kimball methodology is to design data warehouses using an iterative and incremental process, which aims to build a robust and flexible data warehouse over an extended period of time. This process begins with identifying key business processes and data sources, and continues with building dimensional models, establishing data integration processes and refining the data warehouse using user feedback.
share post
