Majority of the operational databases are in 3NF (3rd Normalized Form). Data warehouse, on the other hand, uses dimensional models. What really is the key difference between 3NF and dimensional models?

As I understand it’s the degree of normalization. Remember both model types can be presented as ERDs (Entity Relationship Diagram). Normalized models, however, are too complicated for data queries. RDBMSs can’t query a normalized model efficiently.

The following are some texts from the book “The Data Warehouse Toolkit”.

Dimensional modeling addresses the problem of overly complex schema in the presentation area. A dimensional model contains the same information as a normalized model.

If the presentation are is based on a relational database, then these dimensionally modeled tables are referred to as star schema.

If the presentation are is based on multidimensional database or OLAP technology, then the data is stored in cubes.

Dimensional modeling is applicable to both relational and multidimensional database. Both have a common logical design with recognizable dimensions; however, the physical implementation differs.

so I might be wrong by referring to our Asset Management database an operational database. The data can be updated through the client application, but in general, our data updates are managed-load updates, not transactional updates.