New business intelligence software called Oracle Data Integrator helps users cull analysis from the systems of...
various vendors, including IBM, Microsoft and Netezza. Oracle today announced the general availability of the new business intelligence software, the first major upgrade of Sunopsis Inc. business intelligence software since Oracle purchased the Limonest, France-based data integration vendor last year.
ODI is a data integration offering designed to help users make sense out of the massive amounts of information oftentimes stored in various platforms and applications throughout an organization.
Oracle says the integration software lets companies more easily handle initiatives related to business intelligence software, data warehousing, master data management (MDM), business activity monitoring, application migration and consolidation, and service-oriented architecture (SOA).
ODI is optimized to work with Oracle's own database, application server, SOA suite and business activity monitoring software. According to Oracle, the new version adds features that let users support packaged applications from IBM, Microsoft, Netezza, Sybase, Teradata and Trillium.
The software also lets users build data warehouses that can be accessed and analyzed. But some IT industry watchers, like Gerry Brown, senior analyst with Towcester, U.K.-based Bloor Research International Ltd., believe that BI doesn't necessarily need to be all that pervasive.
"You can't say that everyone in a big organization should have BI, because they wouldn't use it and they wouldn't understand it," Brown Fusion Middleware platform, on which all newly acquired and homegrown Oracle applications will run.
"The Sunopsis tool and Data Integrator now are 100% Java-based, so we've been really able to pull this into the Fusion Middleware stack and do deep levels of integration in just a matter of weeks," said Jeff Pollock, senior director of Oracle Fusion Middleware.
Key features, customers and pricing
Oracle says that one of the most important features of ODI is its ELT (extract, load, transform) technology -- not to be confused with ETL (extract, transform, load) technology, which Oracle calls "previous generation."
"With ELT, you can do transformations in the source or the target systems. You don't require a separate ETL server," said Ashish Mohindroo, senior product director for ODI. "You can do the transformation and push out the data to the target source."
The ELT technology therefore gives the system higher performance and greater support for third-party systems, and it can help reduce costs, Mohindroo said.
Another key feature of Data Integrator, according to Pollock, is its ability to support both batch and real-time data synchronization through its change-data-capture technology.
"What change-data-capture does is it allows us to monitor the log files of downstream legacy systems and databases to look for changing data," Pollock said. "When we see changing data, we can now raise these as events to the Business Activity Monitoring tool natively, which can in turn populate a dashboard."
The executives said ODI lowers the cost of developing and maintaining data integration maps by giving users a set of pre-packaged code libraries called Knowledge Modules, which capture highly optimized source and target specific optimizations for major databases and packaged applications.
Oracle says that customers currently making use of ODI include Accenture, Netezza, Nielsen Media Research, Sabre-Holdings, Teradata and Blue Cross Blue Shield (BCBS) of Massachusetts.
BCBS is using ODI to help cull information from an IBM DB2-based data warehouse containing about 1.2 terabytes of information on healthcare providers and claims for about 2.9 million members. In a statement released by Oracle, BCBS said the ODI software helps users extract specific bits of information from the warehouse, build data marts, and conduct analyses to determine the effectiveness of sales and marketing and other initiatives.
Oracle Data Integrator is generally available for $12,000 per database target CPU and $4,000 per database source CPU.