Page 189 - DCAP603_DATAWARE_HOUSING_AND_DATAMINING

Unit 9: Data Warehouse Refreshment – II

the solution notes
BT first selected the Trillium Software System® to aid data quality and record linking in
a central marketing database sourced from millions of customer records. Since then, BT’s
experiences with the solution have seen the Trillium Software System become the corporate
standard for BT enterprise customer data quality.
“We have expanded and extended our use of the Trillium Software System to the extent
that it has become our defacto standard for name and address processing across BT,” said
Turner. “The Trillium Software System is serving BT across tens of millions of customer
records, benefiting hundreds of processes, and is central to the effectiveness of BT’s
customerfocused operations and competitive position.”
The Trillium Software System has been built into CSS, NAD, and Siebel and their many
source systems (as well as other processes), ensuring the quality of customer names and
addresses, enabling source data formats to be standardized, duplicates to be recognized
and handled, and fragmented customer records to be intelligently molded into linked
information. Matches with external data sources from The Royal Mail, D&B, and others
(for name and address and for lifestyle data, for example) can also be made reliably too, in
order to supplement BT’s own information.
“The Trillium Software System is now much more than an application at BT. It’s a key part
of the infrastructure,” said Turner.
the results
BT is finding that strong customer data quality is enabling it to get the most from CSS,
NAD, Siebel, and other initiatives.

9.6 summary

l z This unit has presented an analysis of the refreshment process in data warehouse
applications.
l z You have demonstrated, that the refreshment process cannot be limited neither to a view
maintenance process nor to a loading process.
l z You have shown through a simple example, that the refreshment of a data warehouse can
be conceptually viewed as a workflow process.

l z You have identified the different tasks of the workflow and shown how they can be
organized in different refreshment scenarios, leading to different refreshment semantics.
l z You have highlighted design decisions impacting over the refreshment semantics and we
have shown how the decisions may be related to some quality factors such as data freshness
and to some constraints such as source availability and accessibility.
9.7 keywords

Data Refreshment: Data refreshment in data warehouses is generally confused with data loading
as done during the initial phase or with update propagation through a set of materialized
views.
Data Warehousing: Data warehousing is a new technology which provides software infrastructure
for decision support systems and OLAP applications.
Materialized View: A materialized view eliminates the overhead associated with expensive joins
and aggregations for a large or important class of queries.
Nested Materialized View: A nested materialized view is a materialized view whose definition is
based on another materialized view.

LoveLy professionaL university 183