DW - Data Warehousing & Modeling


The state of a database when all repeated values within the database have been removed by creating more tables. Relational OLTP systems are usually highly normalized. Data warehouses typically are not highly normalized.

Normalizing data is a process defined by E.F. Codd and boils down to structuring the data into separate and multiple tables to avoid redundancy in the data.

Master Data Management

Master Data Management (MDM) is a set of tools and processes that aim to deliver a single clean and consistent view of each master data entity (product, customer, employee, financial accounts, etc.) that exists within the organization


Levels are an implementation of the hierarchies that represent the level of detail viewed in a dimension. As an example, for a time dimension hierarchy, we have levels for year, quarter, and month.

Information discovery

Information discovery denotes the middle section of enterprise Performance Management and Data Services. It allows decision-makers to access, navigate, analyze, format and share information within the organization. It primarily comprises the Business Intelligence platform, the components of query and analysis, as well as reporting and dashboarding functionalities.

End user layer

The End user layer is the user interface on top of the multidimensional structures designed for data access tools. Here, the technical layer (databases & relations) is transformed into a business layer, understandable for the end-user.

Dimensional hierarchy

A dimensional hierarchy denotes how data is organized at various levels of aggregation. An analyst uses a dimensional hierarchy to identify various trends at one level, drill down to lower levels to detect causes for these trends, and roll up to higher levels to see the effects the trends have on the whole business.


This is the process of converting normalized tables again into a de-normalized form. Here, a table may contain redundant information. This is a common technique within data warehousing where star schemas are used to optimize performance. Denormalizing a database consumes more space and slows OLTP performance but improves query performance in a BI environment.

Change data capture (CDC)

Change data capture (CDC) is the process of capturing changes made at the data source and applying them throughout the enterprise. CDC minimizes the resources required for ETL (extract, transform, load) processes because it only deals with data changes. The goal of CDC is to ensure data synchronicity.


S'abonner à RSS - DW - Data Warehousing & Modeling