IBM has an all-encompassing data management product on the drawing board that promises to further the companys efforts to marry structured and unstructured data.
Known informally inside IBM as the "information integrator," or II, the product will combine capabilities from three existing products and add next-generation XML technology, according to Janet Perna, general manager of IBM Data Management Solutions, in Armonk, N.Y.
The product enters pilot testing in the first half of next year and will be delivered in 12 to 18 months, Perna said.
In a sense, II will be a business coming-out party for two of three existing products, which are largely used in the life sciences industries to search huge disparate database and data sources. They are DiscoveryLink and DB2 Relational Connect, informally known as "data joiner."
DiscoveryLink allows scientists to access multiple data sources databases with a single query, while DB2 Relational Connect provides DB2 with native read access to Microsoft Corp.s SQL Server, Oracle Corp.s Oracle and Sybase Inc.s Sybase. The third IBM product—Enterprise Information Portal—is another querying tool, although its unclear if it will be added totally.
"Whats added to these three products over time is XML and greater integration capability. This supports relational, XML and rich-media views of the world," said Jim Kleewein, an IBM distinguished engineer working on the project in IBMs Silicon Valley Laboratory, in San Jose, Calif. "It takes you beyond just integrating data to merging business with IT processes along with structured and unstructured data."
What makes this more than just the sum of the three products together are two pieces of XML technology developed by IBM, according to Kleewein. The first is Xperanto, a code name for an XML foundation technology that supports XML within DB2. The second is an XML query language based on the XQuery technology developed by IBM and under consideration as a standard by the World Wide Web Consortium. XQuery is part of Xperanto.
II supports IBMs belief that XML and relational data must live together instead of the former replacing the latter.
"Its a game changer for IBM," said Kleewein. "We are expanding our business from database management to data integration and management. Customers real problems are not managing databases. Its managing the data assets of their entire enterprise."
"This is a general approach to integrating and federating all forms of information. Its one of the toughest areas of data management and is based on 10 years of research and product knowledge," IBMs Perna said.
Analysts have not been briefed on II but see it as a continuation of what IBM has been doing for some time.
"In a way, it sounds like IBM is rummaging through their inventory of stuff and is trying to make something interesting out of them," said International Data Corp. analyst Carl Olofson, in Framingham, Mass. "It has to hang together as product rather than come across as a loose collection of things that werent designed to work together. If most of them come out of the DB2 family, they probably will work together."
Another analyst discounted the notion that the unstructured world of XML and highly structured DB2 model will collide.
"They all come out of the same DB2 closet. If you look under the covers, theres DB2," said Herb Edelstein, an analyst at Two Crows Corp., in Potomac, Va.
IBM, Edelstein said, has enjoyed considerable success with DB2 Relational Connect, but its too soon to judge DiscoveryLink, which was introduced in April.
"What IBM is trying like everybody else is not just offer raw tools but to do a fair amount of tool integration," Edelstein said. "And it wants to take DiscoveryLink into a broader-based business intelligence environment."