You can build a cube using the Cube Wizard or Cube Editor. The Cube Wizard takes you through the process in a series of steps. Cube Editor allows you to perform some of the steps in your own order. To build a cube using either the Cube Wizard or Cube Editor, you must specify the data source, fact table, measures, and dimensions for your cube.
The data source contains the fact table and dimension tables you want to include in your cube. A data source name identifies a database resource and parameters for its usage.
The fact table contains the measures you want to include in your cube. A fact table is the central table in a schema. It contains the numerical data (that is, measures) of main interest to end users of the cube. A fact table also contains foreign keys that are joined to primary keys in dimension tables.
The measures that you select are the ones that you want to make available to end users. A measure contains numerical data (for example, Sales) viewed and analyzed by end users. Each measure corresponds to a column in the fact table. This column supplies the measure's values.
The dimensions that you select will also be made available to end users. Dimensions are descriptive categories by which the measures can be separated for analysis. In tabular browsers, dimensions provide the column headings, row headings, and subheadings by which the measures are separated and displayed to end users. (In graphical browsers, they provide other types of descriptive labels but with the same function as in tabular browsers.) For example, the measure is Sales, and the dimensions are Time, Location, and Product. End users can separate Sales into various categories of Time, Location, and Product. Time provides headings for individual years and subheadings for months. Location and Product also supply a variety of headings and subheadings.
Each dimension is created from one or more columns in a dimension table. These columns supply the values of the dimension, and produce the column headings, row headings, and subheadings seen by end users.
Each dimension table contains a primary key that is joined to a foreign key in either the fact table or another dimension table.
You can also build virtual cubes, which combine elements of multiple, previously built cubes. When end users browse the virtual cube, they see the combined elements together as if they were in a single cube. One of the advantages of virtual cubes is that their definitions, but not their data, are stored. Thus, virtual cubes require much less storage space than regular cubes.
You can also build linked cubes, which can provide additional flexibility in distributing cube data to end users.