Understanding Multi-dimensionality

With IBM® Cognos® TM1®, you can create multidimensional databases that provide business and finance managers instant meaning from complex, dynamic business models.

To understand multi-dimensionality, consider the example of the Vice President of Sales for a retail company who wants to analyze product sales across a retail chain that operates in the United States and Canada. Each retail store records the unit sales, dollar sales, and discounts for the durable consumer products.

The sales are analyzed by product, scenario (actual versus budget), region, measures (units, dollar sales, and discounts), and week. This example identifies a five-dimensional model. The dimensions identify how the data is organized or how the types of data are tracked.

In TM1, the sales analysis can reside in one or more multidimensional structures called cubes. A collection of cubes forms a database. Each data point in a cube is identified by one element in each dimension of the cube. For example, actual dollar sales of dryers during the second week of January in the Boston store. TM1 cubes must contain no less than two and no more than 256 dimensions.