dcsimg
RealTime IT News

HyperRoll Looks to Bust OLAP Barriers

Software maker HyperRoll Monday unveiled the third version of its flagship platform for ramping up the performance of business intelligence applications from companies such as Business Objects and Cognos.

Mountain View, Calif.-based HyperRoll competes with Netezza in a small niche of the data aggregation market, where the goal is to help BI applications, which inform companies about the performance of their business, get closer to real-time functionality. Analysts say that as BI platforms are becoming increasingly Web-based, customers are beginning to expect their BI software to update on the fly to keep them updated up to the minute.

That's where HyperRoll 3.0 comes in, according to HyperRoll's Rich Ghiossi, vice president of marketing.

"Everybody wants to get to real-time performance," Ghiossi told internetnews.com in a recent interview. "We allow customers to make incremental changes in performance, approaching real-time analysis." Where data aggregation tasks being performed with BI applications might take as much as 18 hours to complete while going through all of the necessary steps, HyperRoll 3.0 cuts the project completion time down to a half hour by working around certain steps, Ghiossi said.

While enterprise resource planning (ERP) tend to undergo an information exchange from the data warehouse to a traditional online analytical processing (OLAP) cube, HyperRoll 3.0 substitutes for the OLAP cube, which tends to take awhile to build and tune. HyperRoll said it speeds up this process considerably to ensure data reaches reporting applications faster than usual.

Ghiossi said one practical example of how HyperRoll works is through Nortel Networks , which uses Oracle Express to aggregate financial data in a process that previously took a minimum of three hours, or 180 minutes.

"Using HyperRoll, the same data is fully aggregated in five minutes and ready for reporting in less than one minute," Ghiossi said. Accordingly, Nortel has freed up server resources for other applications and reduced storage space.

Because HyperRoll's technology ramps up loading by as much as 100 times, Ghiossi said users of OLAP solutions such as Hyperion Essbase and Oracle Express can access large data volumes while drastically reducing the need for more hardware or storage. Moreover, HyperRoll software is transparent to end users, who do not have to learn a new interface or access a new application to improve the performance of their BI applications.

"They are gaining a reasonable amount of traction," said Eric Rogge, vice president and research director at Ventana Research. "What is surprising to me is the significant need for performance acceleration in a market where hardware costs continue to decline. Products like HyperRoll 3.0 are alternative solutions to the paradox of finding a creative way to speed up performance that doesn't involve brute force of the acquisition of hardware."

Rogge told internetnews.com that a lot of the queries for BI intelligence bump up against a large OLAP cube or large database that could result in some latency of query time, or in the time it takes to construct a cube. It also costs about one-fifth of the cost of a competing solution from Netezza.

End-users can also query against the data while the cube is being calculated, which translates into faster, better visibility into company performance, financials and operations. HyperRoll 3.0 also features native access to DB2, support for 64-bit computing and new memory and caching algorithms for improved engine performance.

In addition to Nortel, HyperRoll boasts a customer list with the likes of Kellogg's, MasterCard, Deutsche Bank and GE Financial Assurance. The company sells its products directly but also partners with a number of BI vendors to pair its software with platforms from IBM, Oracle, Cognos and Business Objects.