IBM Reveals Database Software for Indexing

IBM Monday ratcheted up
its database software line with the introduction of a new suite that
automates the process of defining, describing and indexing information
contained in the database.

Called Cube Views (formerly code-named Aurora), the new business
intelligence middleware uses algorithms to automate the creation of online
analytical processing (OLAP) metadata so that it can be
examined from many points of view. OLAP technology lets users compare and
contrast information.

Using OLAP, a user can request that data be analyzed to display a
spreadsheet showing a company’s products sold in a state in the month of
July, compare revenue figures with those for the same products in another
month, and then see a comparison of other product sales in that state for
the same time period.

Armonk, N.Y.’s IBM competes with the likes of Microsoft and Oracle in this
segment of the database market, which Nigel Pendse, author of OLAP Report, said was worth $3.5 billion
in 2002. The attraction of Cube Views, said IBM Director of Business
Intelligence Development Anant Jhingran, is that it makes it possible for
OLAP metadata to be created once and shared numerous times with end users,
programmers and other business intelligence applications during every phase
of data gathering and analysis. Programmers can use metadata to aggregate
data into four-dimensional, or “cube-like” charts, which show data from
different perspectives.

For example, Jhingran told internetnews.com a cube may show product
sales by region, time period, salesperson, profit margin and marketing
campaign. With Cube Views, one department can use a particular application
to look at their product from a historical perspective, while another
department can use a different set of tools to predict future sales for
another product or region. End users can then use a variety of query and
reporting tools from different vendors that best suit their particular
analysis needs.

Redmonk Senior Analyst Stephen O’Grady said the moe was part of a trend in the database software market where data transformation and reporting features are added to the core database. He cited Microsoft’s Yukon as another product that has added this type of functionality toits reporting and analytics embedded functionality.

“This announcement is also well timed given the importance the BI tools carry within many enterprises right now, and the initiatives that are currently underway with many BI customers to extend their BI data to new employee segments as well as making it more relevant, more digestible, and more insightful,” O’Grady told internetnews.com.

According to Jhingran, the automation of this process makes Cube Views
another piece of software in the large e-business on-demand strategy for
IBM, a field consisting of many players, including HP, Sun Microsystems,
Veritas Software and computer Associates.

Rather than go against the grain of leading business intelligence
specialists, Jhingran said 15 vendors, including Brio, Crystal Decisions,
Cognos, Microstrategy, Informatica and Business Objects, are supporting the
new IBM tool because the technology is standardized in the database and is
complementary to their offerings. Jhingran said this is something that
Microsoft and Oracle do not provide. Rather, they opt to vest their
interests in their own front-ends for OLAP.

“It’s certainly a logical progession that database makers continue to enhance the value add that their RDBMS products offer; what is a bit more difficult, however, is accomplishing this without stepping on any partners’ toes,” O’Grady said. “IBM seems to have managed this successfully, however, because the partner support for the product is excellent.”

Charles Norman, director of business development at Embarcadero Technologies
discussed his company’s OLAP partnership with IBM for Cube Views.

“Metadata management and integration can be a tricky problem, especially
when supporting multiple constituencies with unique business and technical
requirements,” Norman said. “The hub and spoke architecture of IBM’s
Metadata Management platform enables seamless sharing of metadata across all
the leading BI and ETL vendors -this is a huge benefit for our DT/Studio
customers because, right out of the box, it provides metadata integration
with all the major BI vendors tools. We are thrilled to see this type of
database infrastructure technology made available in the market.”

Other features of Cube Views include a “performance advisor” that automates
the information gathering and tagging process and, because data is suited
for certain query methods, third-party business intelligence applications
can work faster. Cube Views also features a XML-based Web Services interface
that yields answers to queries in a more accessible format.

Jhingran said the IT world can expect more feature-rich business
intelligence products down the road from IBM, but would not delve into
specifics. However, he reiterated that IBM’s belief is that business
intelligence is best leveraged in a single version in and around the data
warehouse, providing opportunities for greater scalability and expansion of
data management and analysis.

Cube Views, slated for a June 27 release, will be $7,500 per CPU.