In-Depth
IBM's Information Integrator Update Packs 100 New Features
“Masala” to offer search and data-access enhancements, and brings the product closer to being able to manage both structured and unstructured data
Since when is a beta software announcement newsworthy? When it's about the follow-up to a product (IBM Corp.’s DB2 Information Integrator) that helped jumpstart the Enterprise Information Integration (EII) market.
Because Big Blue claims that its first open beta of the next version of DB2 Information Integrator (code-named Masala) boasts at least 100 new features, including search and data access enhancements that make it worth a closer look.
EII is a composite technology that pulls together information from structured and unstructured data sources. EII actually encompasses a variety of different data integration technologies, including ETL, EAI, workflow, and collaboration.
Last year, IBM announced two versions of its EII offering, one (DB2 Information Integrator v8.1) designed for structured data and another (DB2 Information Integrator for Content v8.2) for unstructured content. The goal, says Laura Haas, distinguished engineer and senior manager of IBM’s Information Integration Development effort, is to eventually deliver a single solution, DB2 Information Integrator, for managing both structured and unstructured content. When it appears, Haas says, Masala will constitute an important step toward that goal.
“This year, our focus is very much on bringing the two versions together and expanding the reach of what I would refer to our more structured offerings to be able to reach content data,” she explains. “What we’re trying to move towards is a single platform so in the future we’ll be going from two separate products really to one. With this announcement, we’re really making it possible for a lot of the integration that needs to occur.”
IBM shipped DB2 Information Integrator for Content several months after delivering a version of the product for structured data. Haas says the same will probably hold true for Masala. “I think you’ll see the content management portion come out a little bit later.”
If its name alone isn’t a sufficient tip-off, DB2 Information Integrator is closely tied to Big Blue’s DB2 Universal Database. Not surprisingly, then, IBM’s next-generation EII offering is also designed to exploit enhancements in the next version of DB2, code-named Stinger. One such improvement—integration with Big Blue’s WebSphere Business Integration (WBI) platform—helps facilitate near-real-time access to PeopleSoft, SAP, and Siebel application data, she says.
“We are inheriting [an expanded set of application tooling] from Stinger, so we actually are exploiting our WebSphere Business Integration family of products. They have adapters to packaged applications such as SAP, PeopleSoft, and Siebel, and we’re making it possible to go through information integrator to those sources,” Haas indicates.
Stinger will cement the integration between WBI and DB2 Information Integrator, when it appears. “The sort of real-time federated access to data sources … is more tightly tied [to Stinger] in that we do use the DB2 capabilities in the base,” Haas confirms.
One new capability that’s not tied to Stinger is a revamped search facility for both structured and unstructured data alike, says Haas. “Our new search capability … gives our customers an alternative way to look for data, [through] a very familiar keyword-based search capability, so that they can search both structured and unstructured data, their portals, their back-end data sources, and so on. These are very scalable capabilities, and they’re really aimed at the overall enterprise for search.”
As of this Masala beta milestone, IBM still hasn’t incorporated programmatic support for the long-awaited XML Query (XQuery) standard. Haas says this isn’t surprising, however, because a finished XQuery standard probably won’t be approved before 2005. Instead, she notes, Big Blue has built support for a similar approach, called SQL XML, into Information Integrator: “What we have in Information Integrator is something called SQL XML, it’s an extension to the XML Standard that does a lot of published data in the XML format, but it’s something that the ISA and ANSI did to extend support for XML.”
Haas reiterates a promise the company made last year: “We will have XQuery support, but we’re developing it as the standard settles down, so hopefully we can release it at the same time as the standard.”
From what IBM has seen, Haas says, customers aren’t beating down the door for XQuery anyway: “We don’t see customers in a hurry to adopt a language that can change dramatically.”
About the Author
Stephen Swoyer is a Nashville, TN-based freelance journalist who writes about technology.