Analysis Services provides new user interface wizards, dialog boxes, and editors to help you quickly perform data mining administrative tasks such as building data mining models and incorporating the results in OLAP cubes. For more information, see xxx.

MDX Extensions for Data Mining

MDX syntax has been extended to provide data mining capabilities in connection with OLAP cubes. For more information, see Mining Model extensions to MDX.

Build Data Mining Models with Data Transformation Services

A new DTS task is provided that you can configure to develop and train data mining models automatically. For more information, see Automating and Scheduling Administrative Tasks.

Data Mining in Client Application

• Analysis Services client applications can use data mining algorithms to discover information in OLAP cubes by creating data mining models and virtual cubes. For more information, see Using PivotTable Service for Data Mining.
• Dimension Enhancements
• This release of Analysis Services adds significant functionality to multidimensional OLAP analysis with a number of new dimension types, features, and improvements.

Parent-child Dimensions

A new parent-child dimension type supports hierarchies based on parent-child links between members in columns in a source table. Such hierarchies represent structures that include organization charts and part assemblies. For more information, see Parent-Child Dimensions.

ROLAP Dimensions

Extremely large dimensions can now be accommodated using the ROLAP storage mode. The dimension data remains in the database table and is not subject to the size limitations of MOLAP. For more information, see Dimension Storage Modes.

Write-enabled Dimensions

The members of write-enabled dimensions can be updated through OLAP Manager and client applications that support dimension write-back. Roles are used to control dimension write access by client applications. For more information, see Write-enabled Dimensions.

Changing Dimensions

The new changing dimension type permits more types of changes than other kinds of dimensions without requiring the dimension to be fully processed after changes. This increases the availability of cubes to client applications. For more information, see Changing Dimensions.

Dependent Dimensions

Dependent dimensions permit improved aggregation design optimization by using knowledge of non-intersecting member combinations to reduce storage requirement estimation in the design algorithm. For more information, see Dependent Dimensions.

SQL Server 2000 Preview 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19