IBM makes databases more intelligent

Computing giant IBM has continued the trend of packing more business intelligence functionality into databases with the release...

Computing giant IBM has continued the trend of packing more business intelligence functionality into databases with the release of DB2 OLAP Server 8.1.

The new software pulls together what IBM calls "hybrid analysis" and Online Analytical Processing (OLAP) with data mining technology. Hybrid analysis is the combination of relational and multidimensional data, such as OLAP cubes, according to Jeff Jones, senior program manager of IBM data management solutions.

"In the past you've had to choose between OLAP and relational data," Jones said. "Now customers have control over how to weave together relational storage and multidimensional OLAP data."

Jones said that users can more easily find answers to intricate queries as well as mine for data summaries or broader trends within the data. The newer model also blends the scalability of the relational data engine with the higher-performing OLAP, Jones added.

Being able to run data mining on both OLAP and relational data boosts the performance of analytics, said Mike Schiff, vice-president of business intelligence and e-business at market research firm Current Analysis.

"Allowing users to combine both relational and multidimensional data will increase the utility of data mining," Schiff said. "The ability to report from both relational and multidimensional data also makes accessing information easier for users who could not care less where it is stored."

Two of IBM's database archrivals, Oracle and Microsoft, have been working to increase the strength of their OLAP capabilities. Microsoft started the trend when it pulled OLAP into its SQL Server 7.0.

More recently, Microsoft has been involved with the XML for Analysis Council. The organisation was formed to create a protocol that the participating companies hope will become an industry-standard messaging interface for business intellignence. SAS Institute and Hyperion Solutions are also behind Microsoft's efforts, though both companies are supporting Java OLAP (JOLAP) as well. JOLAP, as the name implies, is a Java-based specification for creating and managing data in OLAP servers and is currently in development within the Java Community Process. IBM is also behind JOLAP.

Oracle, for its part, integrated OLAP into its 9i database. Previously, it was available as its OLAP tool Express. The company early last month ended Express' life as a stand-alone product and rechristened it Oracle OLAP with plans to make it available as an add-on to Oracle 9i Release 2, due later this month.

Read more on Business intelligence software