Export (0) Print
Expand All

Object Architecture Overview

SQL Server 2000

Microsoft® SQL Server™ 2000 Analysis Services provides a variety of objects to help you implement an online analytical processing (OLAP) or data mining solution with a data warehouse. This topic includes descriptions of the available objects.

The main object of OLAP is the cube, which contains the current analytical data of interest to end users. To support the questions that end users ask, cubes organize data into dimensions and measures in a multidimensional structure. For example, consider the question, "What was our total sales of hardware in the northwest region in the first quarter of this year?" A cube of data that can answer this question includes three dimensions and one measure:

  • The Product dimension, which contains a hardware category

  • The Geography dimension, which contains the northwest region

  • The Time dimension, which contains the first quarter of this year

  • The Sales measure, which contains quantitative numerical data that can be summarized

Whereas OLAP allows you to perform aggregation analysis on current or past data, data mining actually allows prediction analysis to be performed based on current or past data. Instead of considering the question posed earlier in this topic using OLAP, the question, "What will our projected total sales of hardware in the northwest region be for the first quarter of next year?" can be asked and answered with data mining. The main object of data mining, the data mining model, provides a framework to store learned knowledge from your data, such as probability and distribution information, created from existing data to predict the behavior of new data. This, in turn, can be given new data for analysis, to predict expected values for a given case based on patterns and rules discovered in past data.

Object Hierarchy Diagram

The objects used to support OLAP and data mining are represented by a object hierarchy, used to maintain the complex relationships between the various objects, such as cubes, dimensions, and data mining models, that define Analysis Services.

The following diagram shows the positions of the objects within the Analysis Services object hierarchy. Some objects appear in multiple places within the hierarchy.

Analysis Services object hierarchy

This topic describes the administrator's view of the object model. The programmer's view is somewhat broader and more complex. For more information, see Decision Support Objects.

© 2014 Microsoft