ETL Process for the Model Builder DTS Task
The Model Builder DTS task allows users to schedule rebuilds of models, ensuring that models will be built even if failovers occur during the build, such as database or Predictor service failures. The Model Builder DTS task should be run after manually configuring the Predictor resource. For more information about manually configuring the Predictor resource, see Configuring the Predictor Resource.
Each time an analysis model is built, it is based on the data currently in the Data Warehouse. It is important to rebuild analysis models when the data in the Data Warehouse changes, such as after new data is imported into the Data Warehouse.
You can use the Model Builder DTS task to rebuild analysis models for a specified site or for all sites associated with a specified Data Warehouse. Multiple models can be built by adding multiple instances of the Predictor DTS task to the package. A single task builds only one model.
See Also
Running the Model Builder DTS Task
Scripting the Model Builder DTS Task
Running the Predictor Resource
Copyright © 2005 Microsoft Corporation.
All rights reserved.