|
Philosophy
Business intelligence is an
integrated framework of an information system that provides the
business with a complete view of business activities and
operations. Data warehousing is frequently used for decision
support within a company or organization and provides the
organization with organized, updated and coordinated data that
provides a view of relationships between information obtained from
different sections of the organization.
Our
data warehousing design involves generating, manipulating and
mapping models with business intelligence focus on an overall view
through one front regardless of data storage style, number and
locations. The models are physical, logical, and conceptual
representations of the business data and end-user information
requirements.
Creating
a data warehouse requires the mapping of data between source to
target models, capturing the details of the transformation in the
data warehouse by utilization of tools that support these
activities which are known as data warehouse mapping and
transformation design tools.
By
definition a "Data Warehouse" is a very massive database
that frequently provides access to all of an organization's data
(information). Data could be distributed over several databases on
multiple computers and may contain information from numerous
sources in different formats and varieties. The user data access
to data source (regardless of location, format, type or hardware
etc) should be transparent.
Data
warehouses also contain information about how the warehouse is
structured (it has data about the data stored in it) and
organized, where the information is and the relations between the
data.
Architecture
1) Data collection
and injection
Gathering
data from a wide variety of systems is a challenge, to simplify
this task, it is best to make some form of mapping tool or to
find an off the shelf one to fetch, transfer and load the data
from the various applications. Such tools identify the
information contained in data source systems and identify where
that information needs to be loaded into the data warehouse.
-
Extract
Function to get data from heterogeneous data sources
and different systems.
-
Transform
Function of transforming extracted data to be suitable
for the warehouse.
-
Load
Load
the data to the warehouse
2) Metadata
Management
Tools
necessary to locate, alter, create, and manage metadata
objects--including technical, business intelligence, and
operational metadata within the data Warehouse.
-
Metadata
search
Is required to enables you to perform searches on technical,
operational, and business intelligence metadata components.
-
Modification
handling
Assist you to maintain consistency when developing new
components or reuse previously entered ones.
Covers data names, table names and
what is in the database (information about the data kept in the
database) etc.
3) Impact
Analysis
Facilities that enable you to determine the
interdependencies that a change to one data warehouse object may
have on other related objects.
Example:
A change to a data warehouse table may require change to the
related tables, table maps, or data maps.
4) Information Access
Infrastructure
for 360° information analysis to manage and explore:
Example:
Use scorecards to view and manage strategic initiatives
through client software or web portal. Plugged in analytical
applications are used to provide immediate insight into data
assets.
|
