Dimensional Modeling describes a process of designing a data warehouse assuming that the data are organized as dimensions and facts.
Although the design process is also applicable for OLAP tools, this is a design technique based on the assumption that the data model will be deployed on relational database.
Dimensional modeling is for addressing the reporting requirements. The data used in answering typically business questions are categorized into two groups: fact and dimension.
A typical business question is like “how many computers we sold by regions last quarter”.
The sales figure is the metric. The metrics are facts.
There are the dimensions involved here:
Dimensions are used in filtering, grouping and sorting the answers.
From relational modeling perspective, dimensional modeling produces the data model design with denormalization.