News

Informatica rolls real-time data integration system

• By Jack Vaughan
• June 11, 2002

[June 11, 2002] - Business software maker Informatica in recent years saw its customers’ analytic implementations moving from the batch to near-real-time computing realm. This led the company to launch products that support what Informatica describes as real-time enterprise data integration solutions.

 The new products -- highlighted by Informatica’s PowerCenterRT -- take in ‘trickle’ data from messaging middleware feeds, quickly process this data by applying analytical filters, and then populate a target application to create what might be described as a combination operational and analytic system.

 Developers seeking to work with the new system should be prepared to incorporate MQ, or Tibco EAI development paradigms into their data integration mix. Informatica also offers an SDK for integrating feeds, with WebMethods and JMS messaging means anticipated later this year.

 “The big shift we see is in the applicability of analytics systems in the operational world,” said Girish Pancha, senior vice president and general manager for Informatica’s Platform business Unit.

 Enterprise analytical software users started updating their stores weekly, then daily, and now are updating multiple times a day. “They are asking for ‘real time.’ It’s no longer about batch,” Pancha said.

 The PowerCenterRT engine, he said, processes data quickly, treating it in a different way than Informatica has treated it in the past. The system supports an active metadata approach that maintains data lineage, allows for impact analysis, and is broadly extensible for integration with other enterprise applications and systems.  

 “Before, we assumed [the presence of] larger chunks of data that didn’t need to be processed immediately,” said Pancha.

 “This is a fundamental change in the way our engine works. Now, it’s a nonstop engine,” he said.

 Infrastructure and interoperability are increasingly important to IT shops, said Pancha. “They’ve adopted EAI for one problem, and data integration technology for another.”

 If Informatica’s bets prove right, these two approaches may be combined.