Database Modeling Overview
IBM® SPSS® Modeler Server supports integration with data mining and modeling tools that are available from database vendors, including IBM Netezza, Oracle Data Miner, and Microsoft Analysis Services. You can build, score, and store models inside the database—all from within the IBM SPSS Modeler application. This allows you to combine the analytical capabilities and ease-of-use of IBM SPSS Modeler with the power and performance of a database, while taking advantage of database-native algorithms provided by these vendors. Models are built inside the database, which can then be browsed and scored through the IBM SPSS Modeler interface in the normal manner and can be deployed using IBM SPSS Modeler Solution Publisher if needed. Supported algorithms are on the Database Modeling palette in IBM SPSS Modeler.
Using IBM SPSS Modeler to access database-native algorithms offers several advantages:
- In-database algorithms are often closely integrated with the database server and may offer improved performance.
- Models built and stored “in database” may be more easily deployed to and shared with any application that can access the database.
SQL generation. In-database modeling is distinct from SQL generation, otherwise known as “SQL pushback”. This feature allows you to generate SQL statements for native IBM SPSS Modeler operations that can be “pushed back” to (that is, executed in) the database in order to improve performance. For example, the Merge, Aggregate, and Select nodes all generate SQL code that can be pushed back to the database in this manner. Using SQL generation in combination with database modeling may result in streams that can be run from start to finish in the database, resulting in significant performance gains over streams run in IBM SPSS Modeler.
If connectivity is enabled, you see the option Server Enablement in the License Status tab.
See Connecting to IBM SPSS Modeler Server for more information.
For information on supported algorithms, see the subsequent sections on specific vendors.