Browsing Analysis Services OLAP Cubes with Excel - Microsoft Business Intelligence Tools for Excel Analysts - page 247

➤ Consistent results: OLAP databases contain only snapshots of data. That is, the data in an
OLAP database is typically historical data that is read-only, stored solely for reporting purposes.
New data is typically appended to the OLAP database on a regular basis, but the existing data
is rarely edited or deleted. This allows you to retrieve consistent results when building your
reporting solutions.

Understanding OLAP Cubes
The backbone of an Analysis Services database is the OLAP cube. A cube can be thought of as an
analytical matrix consisting of dimensional coordinates where each coordinate contains a calculation
for every unique intersection. Think of a PivotTable where the data fields in the Rows area and
Columns area intersect to calculate the figures in the Values area. In an OLAP cube, the data fields are
called dimensions and the calculated values are called measures.

Understanding dimensions and measures
For an analytic engine like Analysis Services to be useful, it must work with source data that can be
measured or quantified in some way, possibly in many different ways. This means that the source
data must contain numeric fields that can be summed or otherwise computed using a mathematical
aggregate function such as Sum() , Count() , Min() , or Max() . Quantitative columns are mea-
sures, and there are countless examples of measures in the real world: revenue, cost, quantity sold,
and employee count, to name a few.

Descriptive data provides context, meaning, and structure to quantifiable data. Descriptive fields are
dimensions . There are countless examples of dimensions, but here are a few: date, territory, product,
customer, and sales rep.

Dimensions can get more complicated than measures. This is because dimensions not
only provide context and meaning to measures, but also behave like glue in terms of
making the analytics system feasible. For example, dimensions can relate to measures,
to other dimensions, or to themselves.

Note

Understanding hierarchies and dimension parts
While relational databases are primarily concerned with relationships for data integrity, storage, and
retrieval purposes, Analysis Services uses the relationships as a set of instructions that indicate how
each node in a dimension relates to other nodes in order to form a whole structure. Put another way,
Analysis Services uses data and data relationships to build hierarchies.