InformationWeek: The Business Value of Technology

November 20, 2000

Printer ready

Oracle Aims To Bring Data Warehousing To Net Speed

Oracle9i database will include data transformation, personalization, and OLAP

By Rick Whiting

More on data warehousing:

Oracle9i: It's all about the data (11/6/00) Ramping Up Data Warehouses For Timely Information (11/6/00) Network Computing: Hannaford's Data Warehouse: A Smarter Way To Stock Shelves (11/13/00)

Send Us Your Feedback

n a bid to bring real-time data warehousing closer to reality, Oracle this week will detail plans to build data transformation, personalization, and online analytical processing capabilities into its next-generation Oracle9i database. The vendor also will unveil a new release of its Oracle Warehouse Builder software for automating the design and deployment of data warehouses.

Data warehouses usually are separate from operational systems. Typically, information is transferred from operational systems to the data warehouse daily, weekly, or monthly; analysts then study it for sales, marketing, and business-performance purposes.

But that's too slow for most Internet economy applications. E-commerce companies, for example, want to improve their Web sites' personalization capabilities by combining clickstream data generated by site visitors with warehoused historical information for real-time analysis.

By building business-intelligence and data-transformation capabilities directly into the core database, Oracle9i, available this spring, will integrate data from operational systems and data warehouses for real-time analysis, the vendor says. That should let users such as call-center employees and reservation agents make better-informed customer-related decisions. Earlier this year, NCR Corp. introduced a similar concept called "active warehousing."

"The idea is to close the loop between the decision-support system and the operational or transaction system," says Lou Agosta, a Giga Information Group analyst. On the downside, conducting real-time analysis within an operational database could slow the system, he says.

The new features may make up for potential slowdowns. Oracle plans to embed OLAP technology into Oracle9i and create a Java component-based architecture and a Java API. Database owners and third-party software vendors would use the API to develop analytical applications to take advantage of the OLAP capabilities. Oracle9i also will include built-in data-extraction, transformation, and loading technology, as well as an embedded real-time recommendation engine for personalization.

Oracle also will add data-mining technology to Oracle9i release 2, a move that IBM, Microsoft, and NCR have already made. Says Agosta, "The strategic direction here is to drive data mining into the database."

Back to This Week's Issue
Send Us Your Feedback
Top of the Page