Business Intelligence Tutorial | Dimensional modeling and metadata

Go back to Tutorial

Free Business Intelligence Practice Tests - BI ETL

In simplest language, metadata is referred as “Data About Data”. The metadata is of two type, first Structural Metadata and another one is Descriptive Metadata. The term “Data about Data” in not applied in case of structural metadata. As the Structural Metadata hold specification about structure and design of data. At this movement it, doesn’t contain any data.

On the other hand the Descriptive Metadata contains the information about the data present in the database or simply in application. The descriptive metadata contain the useful information about the data present in the database or data warehouse. For this reason it is known as “data about data” or sometime also referred as “content about content”.

Dimensional Modeling (DM) is used to design a data warehouse. In simple terms Dimensional Modeling is used to support end-user queries. DM is not only applied to relational databases but can also apply to non-relational databases. The DM use the two main concept first one is “FACT” and other one is “Dimensions”. Fact also refers as Measures which are typically numeric values. Dimensions are refers context, which are group of descriptor which defines facts. Every Phase of Dimensional Modeling results in some metadata which can be used to identify what exactly present in data warehouse.

In dimensional modeling, the facts are encapsulated with star-like structure of dimensions. There are four step used to build DM based data warehouse.

  • Identify the business process.
  • Declare the grains.
  • Identify Dimensions, and
  • Identify facts.

The benefits of DM are following

  • Performance: The DM model provides end users quick response to their query.
  • Understandability: The end user can easily understand as the information is combined into category of dimensions.

 

Go back to Tutorial

Get industry recognized certification – Contact us

Menu