A Semantic Layer is a translation layer that sits between the Data Warehouse (with Data Marts) and the end user.

A Subject Matter Expert ('SME') is frequently involved in defining the translations, which can be stored in an online data Dictionary.
 
This means that Users can be 'Self-Service' and define their Reporting requirements using common familiar businessterms. 

Wikipedia - 
A semantic layer translates Data Modelling terminology into business-friendly representation that helps Subject Matter Experts (SMEs) and 
end users to access data on a self-service basis with using common business terms. 

In other words, it maps complex data into familiar business terms such as product, customer, or revenue to offer a unified, consolidated 
view of data across the organization.

This shows the four layers in a typical Data Architecture :-







 © DataBase Answers Ltd. 2014