Database administrators (DBAs) can expect the new version to make their lives easier with an automated query optimisation utility, said Jeff Jones, senior programme manager of data management solutions at IBM.
DBAs can already run a utility that gathers statistics on data geography and optimises subsequent query requests, but the next release will perform these functions in real time.
"In essence the database will maintain an understanding of how the data is laid out so that queries are automatically optimised for performance," Jones said.
According to Jones, the new features reflect IBM's move to extend the reach of its grid computing initiative.
"It makes sense for IBM to introduce self-managing functions in products such as DB2," said Galen Schreck, an analyst at Forrester Research. "It allows them to test some of the components of grid computing in a controlled environment."
On the XML front, Jones said DB2 would offer more support for XQuery, one of the emerging standards for querying XML data.
"We will demonstrate how XQuery can talk to DB2 later this month," Jones said. "This is an intermediate step for us in XML."
Further down the road, Jones said IBM also plans to offer a separate, non-relational XML data repository, part of another larger IBM initiative dubbed Xperanto.
Meanwhile, IBM's plans represent a marked contrast the strategies of competitors Oracle and Microsoft, which are sticking to the relational database paradigm.
Jeff Ressler, lead product manager for SQL Server at Microsoft, believes XML data will be very compatible with relational databases.
"As XML becomes more ubiquitous, it will become more integrated with the RDBMS [relational database management system] and also with everything else," Ressler said.