IBM demos new data-access technology

The company has posted a demonstration of the data-access technology, developed through an internal project called Xperanto and bound for IBM's DB2 database server software later this year. The project uses a combination of technologies to allow people to search relational databases, XML (Extensible Markup Language) documents, flat files, spreadsheets and other sources of information as if they were a single database, IBM said.

The Xperanto project, begun a little over a year ago, uses XML, a Web standard for data exchange used extensively in Web services applications; XQuery, a proposed standard language for finding information in XML-based documents; and text search capabilities and Web services techniques to demonstrate IBM's future data integration plans, said Nelson Mattos, director of information integration initiatives at IBM's Santa Theresa Research facility in San Jose, Calif.

All of the technologies, except XQuery, are available in the current version of DB2, Mattos said.

The Xperanto demonstration shows how a newly merged bank and financial services company could use XQuery, in conjunction with DB2, to search for information in multiple databases and use Web services to make that data available to customers or sales representatives.

Mattos said IBM is developing the technology in parallel with the World Wide Web Consortium's efforts to standardize XQuery.

"We're doing a preview of the technology so customers can understand how to use it," Mattos said, adding that IBM will deliver some of the Xperanto technology in the second half of this year.

Analysts said the trend toward easy access to all business data through database servers is catching on. Other software makers, notably Microsoft and database market leader Oracle, are working to solve the same problem: how to quickly find and access data stored anywhere on a network.

"This is the beginning of an era of an expanded role for database-management systems," said Teri Palanca, an analyst with Giga Information Group. "The idea is to access all of your data sources from one point. And you get the benefit of truly blind programming against data sources that are very diverse."

Besides easier searching and access, the new approach from database makers will make it possible to back up and manage many sources of business information from one place, Palanca said. That could help safeguard vital information and save money by eliminating redundant systems management.

"Especially now, when companies are resource-constricted, if you can streamline that process, so much the better," Palanca said.

Microsoft is developing a version of its SQL Server database, code-named Yukon, to integrate data from multiple sources and let people query that data as if it were stored in one place. The same technology is expected to find its way into the next major release of Windows, code-named Longhorn, in the form of a new file system. Longhorn is expected to debut next year.

Oracle has also discussed using its database server as an integration point for diverse sources of data.

Palanca said customers of the database software companies aren't using the new technology yet since so much of it is still in development. "Nobody is stepping in yet," she said. "People are still fiddling with XML and will have to know how to debug and manage these systems."