Business intelligence software provider SAS Institute Inc. said it intends to expand its efforts to integrate its analytics framework and applications directly into several relational and non-relational database management platforms commonly used by manufacturers to store and manage transactional data.

The move will allow manufacturers to more quickly create, implement, and run analytical models used, for example, to analyze large amounts of warranty information to understand the underlying product defects, said Scott Van Valkenburgh, SAS director of alliances, in an interview with Managing Automation.

Traditionally, in order to use analytical tools such as SAS’ data mining software, users would download large amounts of data from transactional database management systems such as Teradata’s into SAS’ environment. The data would then be analyzed, and results passed back to the transactional database.

That approach, however, requires lots of time for data to be shipped back and forth between the transactional database and the analytics environment. It also requires data migration code that must be validated and tested whenever the SAS or database environment changes.

Integrating the analytics framework and applications directly into the transactional database eliminates much of the testing and speeds up the creation and execution of analytical models, Van Valkenburgh said. “With this approach, you can develop models in one day versus up to six months with the traditional approach.”

Making the integration of transactional databases and analytical tools possible, Van Valkenburgh said, is the advent of new database technologies, including user-defined functions that allow for the compilation and execution of existing code directly in a commercial database management system.

SAS has been working with Teradata on integrating its analytical framework and applications into the Teradata database. SAS announced plans to expand those efforts to other database management system vendors, including Netezza, IBM, Hewlett-Packard, Aster Data, and Greenplum.

SAS and those database vendors will jointly market solutions based on embedded SAS analytics, Van Valkenburgh said.