Information Technology Reference
In-Depth Information
2.2
Data Warehouse and Data Mart
In data warehousing, stores of information data are created. Information data is
extracted from the operational data (raw data) and then cleansed, transformed and stored
in a separate database (the warehouse) and thematic data marts (database concerned with
a specific area of interest). Without impacting the operational databases, analysts can
query the warehouse by means of a catalogue (a dedicated database) containing metadata
that allows users to automatically identify and locate useful data available somewhere
within the organization.
Once again, these warehouse solutions have improved user access to data and made a
step forward to knowledge. However:
x These solutions require to be developed within IT departments where there is a
risk of a lack of understanding of the analyst's specific needs
x Customized query/reports, spreadsheets and graphical applications have been the
main tools utilized for analytical purposes while “ad hoc” analysis has been
severely limited when under constraint
x Connectivity and integration to external data sources have been quite impossible
in addition to discrimination between reliable and untrusted sources
2.3
Business Intelligence
Business Intelligence (BI) was added to data warehouse applications to fill the
accessibility and integration gap, to provide sophisticated analytical processing tools, to
improve the process of transformation from data to information, to allow a rapid delivery
and presentation of information and to extend the quality and value of knowledge
available to analysts, executives and decision-makers. Basically, BI aims to achieve the
best quality knowledge from all collectable data in order to gain a competitive advantage
for enterprises.
BI applications are decision support tools that enable real-time and interactive access
and analysis of highly reliable information aimed at not only quickly identifying
problems and opportunities but also preventing loss of knowledge resulting from the
massive accumulation of data which is neither accessible, reliable or integrated with
other real-time sources. BI applications are also referred to as knowledge management,
data mining, multidimensional analysis or On-line Analytical Processing (OLAP).
BI applications allow organizations to become proactive and information-agile:
x providing customizable queries and reports for specific issues
x empowering analysts to ask intuitive and complex “ad hoc” questions
x gaining a better understanding of transactional and operational information by
“drill down” and “slice and dice” functions
x strengthening accessibility and the integration of dynamic information collected
“on the fly” (e.g., Enterprise Information Portal-EIP) from collaborative
processing (documents, e-mail, spreadsheets,Web pages, etc.) and from decision
processing supplied by a wide range of corporate resources such as data
warehouses, data marts, on-line transactional processing (OLTP) and strategic
applications (ERP, CRM).
A large number of applications belong to the family of the so-called BI, each solution
aiming to fulfil the specific requirements of an enterprise. Figure 1. represents the flow of
Search WWH ::
Custom Search