This week, IBM will deliver its long awaited DB2
Information Integrator, codenamed Masala, whose sophisticated
search engine will let corporate users dig out and analyse data
across a heterogeneous range of information data
stores.
IBM hopes to offer corporate IT shops better search capabilities
than competitors such as Microsoft, Oracle, Yahoo and Google by
collecting unstructured data from a broad range of formats
scattered across and outside the enterprise.
"With a lot of companies starting to make noises about investing
in internet search capabilities, IBM has been working somewhat
under the radar on enterprise search that can take in more than
HTML documents," said one source. "The company is hoping to come at
this from a different angle."
In one fell swoop Masala could solve three problem areas in
managing enterprise-level searches: the rapidly expanding universe
of data, the growing variety of mostly unstructured data, and the
patchwork of databases and data stores.
Some analysts believe Masala is a step in the right direction to
solving some of the thornier problems many users are facing in
collecting and analysing unstructured data.
"What IBM will be able to do is offer a federated data model
that brings together a number of disparate sources in one place so
users can search, index and retrieve data without writing to
individual data sources as they might have had to in the past,"
said Stephen O'Grady, senior analyst at Redmonk. "It is a fairly
significant step up."
All this year IBM officials have been talking about what they
see as a lush opportunity in the area of business intelligence. IBM
has redoubled its efforts to pursue that opportunity through a
co-ordinated effort that spans its $14bn (£770m) software
division.
Masala has been at the heart of that effort to help redefine
business intelligence. It is designed to let corporate users create
a virtual database by collecting information seamlessly across the
enterprise from customer service records, e-mails, tables of
numbers, photos and other forms of information and view them all as
if they were in one location.
It will also let IBM compete with traditional enterprise
information integration suppliers such as Composite Software,
MetaMatrix and BEA's Liquid Data as well as enterprise search
suppliers such as Verity, Endeca and Autonomy.
"What IBM will be able to do is sell against either by
leveraging the strengths of both types of capabilities," O'Grady
said. "This sort of functionality really is a weakness among most
major middleware suppliers."
DB2 Information Integrator can cross-reference historical sales
records in data warehouses with real-time sales information or
unstructured content such as supplier contracts. Users can build a
complete picture of product performance by data or supplier that
can then be viewed on a data dashboard or corporate portal.
An IBM official said the company had deployed an early version
of DB2 Information Integrator on its own intranet spanning 300,000
staff and the program was immediately able to handle 80,000 queries
a day.
Ed Scannell and Cathleen Moore write for InfoWorld