Data sets in templates, information objects, and data objects provide easy access to the data fields that contain the data for a report. For example, a data object or a data set named CustomerData can contain customer information fields, such as customer name, address, phone number, and so on. You choose this data set to create a report that lists customers and their contact information.

BIRT data objects and information objects are made up of data sets and sometimes cubes. Data sets classify data fields retrieved from the data source in dimensions, measures, and attributes. These analysis types are specified at the time of designing the information object or data object, and cannot be modified using Report Studio. It is useful to understand these analysis types to use them appropriately in a BIRT design.

Dimensions are data fields that group other data fields. Each dimension field can contain multiple attributes. Attributes are typically associated with a dimension. Table 2‑1 lists the possible attributes for each dimension field.

Measures are data fields that can be aggregated or computed, such as revenue, profit, price, and so on.

A data object can also contain a BIRT data model comprising one or more linked data sets. You can use a data model in the same way that you use a data set. Additionally, you can use a data model to create a cross tab design. BIRT data models are located alongside cubes in a data object. Query optimization in a data model offers better performance than a cube. A data model also provides you with more flexibility in your choice of dimensions and measures.