IBM database integrates diverse data

IBM last week announced software designed to make it easier for companies to integrate and manage business information scattered across the corporation.

Available in beta version, DB2 Information Integrator ties together multiple data types residing in various computer systems inside and outside corporate boundaries. The software lets users access structured and unstructured data, including XML documents, e-mail, multimedia files and even non-IBM databases, and creates a complete view of their information assets, says Nelson Mattos, director of information integration for IBM.

For example, with one query, a user could access relational data in DB2, Microsoft SQL Server and Oracle databases; images stored in a Documentum content management system; Lotus Notes e-mail; Microsoft Excel spreadsheets; and Web services generated by WebSphere Application Server. From the user's standpoint, the data appears as if it came from one place, Mattos says.

Key to the design of DB2 Information Integrator is that it synchronizes distributed data without requiring that it be moved to a central repository - what IBM calls a federated approach to data integration, Mattos says. Rather than force companies to replace existing databases or move their data, DB2 Information Integrator lets users leave their distributed data stores intact. It works across multiple platforms, including Linux, Unix, Windows and mainframe environments, and it works in near real-time unlike typical batch-oriented methods of synchronizing data.

To continue reading, register here to become an Insider. You'll get free access to premium content from CIO, Computerworld, CSO, InfoWorld, and Network World. See more Insider content or sign in.

IBM last week announced software designed to make it easier for companies to integrate and manage business information scattered across the corporation.

Available in beta version, DB2 Information Integrator ties together multiple data types residing in various computer systems inside and outside corporate boundaries. The software lets users access structured and unstructured data, including XML documents, e-mail, multimedia files and even non-IBM databases, and creates a complete view of their information assets, says Nelson Mattos, director of information integration for IBM.

For example, with one query, a user could access relational data in DB2, Microsoft SQL Server and Oracle databases; images stored in a Documentum content management system; Lotus Notes e-mail; Microsoft Excel spreadsheets; and Web services generated by WebSphere Application Server. From the user's standpoint, the data appears as if it came from one place, Mattos says.

Key to the design of DB2 Information Integrator is that it synchronizes distributed data without requiring that it be moved to a central repository - what IBM calls a federated approach to data integration, Mattos says. Rather than force companies to replace existing databases or move their data, DB2 Information Integrator lets users leave their distributed data stores intact. It works across multiple platforms, including Linux, Unix, Windows and mainframe environments, and it works in near real-time unlike typical batch-oriented methods of synchronizing data.

DB2 Information Integrator will be an asset in call center settings, where agents need access to information about customers that resides in multiple databases along with unstructured sources such as e-mail messages and flat files, Mattos says. In addition, financial services organizations could tie together in-house customer bank records with external investment information.

DB2 Information Integrator is the first products of IBM's "Xperanto" research project, which it kicked off in 2001. But it's not IBM's first data-federation software. The company introduced DataJoiner in 1995 for integrating relational data, and Enterprise Information Portal in 1998 for integrating unstructured content.

This combination differentiates IBM from other vendors that offer "enterprise information integration" software, says Philip Russom, research director at Giga Information Group. The market includes small players such as Nimble Technology, Enosys Software and MetaMatrix, along with more established vendors such as BEA Systems - which last year came out with its Liquid Data for WebLogic information integration product - and Vitria, Russom says.

However, these vendors are focused on structured data and haven't tackled unstructured sources, Russom says. "IBM is the only one that has put the two together."

IBM will ship two versions of the software: DB2 Information Integrator and DB2 Information Integrator for Content. The features are the same. The difference is in the programming tools, which are aimed at users with different skill sets.

Both versions will begin shipping by the end of June, according to IBM.

Read more about software in Network World's Software section.