Architectural Components of DWH

Data Warehousing/Mining 1
 What We Learned in Last Lecture
 Overview of the Components

 Source Data Component

 Data Staging Component

 Data Storage Component

 Information Delivery Component

 Metadata Component

 Types of Metadata
Data Warehousing/Mining 2
 Objectives of Today's Lecture
 Understand data warehouse architecture

 Learn about the architectural components

 Review the distinguishing characteristics of data warehouse

architecture

 Examine how the architectural components supports the

flow of data

 Comprehend what technical architecture means

 Study the functions and services of the architectural

components
Data Warehousing/Mining 3
 Architecture
 The various components of DWH together form the
architecture of DWH
 The correct architecture is critical for the success of your DWH
 The architecture is further composed of the rules, procedures,
and functions that enable your DWH to work and fulfill the
business requirements
 Architecture is made up of technology that empowers your
DWH
 It is a comprehensive blue print.
 The architecture defines the standards, measurements, general
design and support techniques
Data Warehousing/Mining 4
 Architecture contd..

 Architecture in three major areas

– Data acquisition
– Data storage
– Information delivery
 We will concentrate on functions, services,
and technical architecture in these major areas

Data Warehousing/Mining 5
Data Warehousing/Mining 6
 Distinguishing Characteristics

 OLTP architecture is created by keeping in

mind the context of operational systems
 DWH architecture is complex, wide and
expensive

Data Warehousing/Mining 7
 Different objectives and scope

 Architecture has to support the requirements for

providing strategic information
 User is interested in obtaining large result sets
 Defining the scope of the DWH is also difficult
– Have to configure the number of data sources
– What are the external sources
– Are you planning to include the private data too
– Data granularity is also kept in mind

Data Warehousing/Mining 8
 Data Content

 “Read only” data acts as primary component

of the architecture
 Your DWH must support the sorting of data
grouped by business “subjects”
 Architecture must support high volume of
data because..
– When we mention historical data stored in DWH,
we are talking about very high data volumes.

Data Warehousing/Mining 9
 Complex Analysis and Quick Response
 Your DWH architecture must support variations for
providing analysis.
 Users must be able to..
– Drill down
– Roll up
– Slice and dice data
– Play with “what if” scenario
 User must have the capability to review the results in
different output options like
– Text
– Tabular format
– Graphical charts
 Your DWH architecture must have capability of
quick response
Data Warehousing/Mining 10
 Flexible and Dynamic

 After even implementation of DWH your

DWH architecture must be felxible and
Dynamic because
– What if something was missed during
development
– User claims to add another dimension in the
DWH
– Business conditions changed
– Business change calls for additional business
requirement to be included in DWH

Data Warehousing/Mining 11
 Meta data Driven
 Architecture should be meta data driven
because..
– Meta data surrounds the entire movement of data
from source to the end users
– It holds data about every phase of the movement,
and, in a true sense, make the movement happen
 Data dictionary of DBMS of the operational
system is just a faint shadow of the meta data
in DWH

Data Warehousing/Mining 12
 ARCHITECTURAL FRAMEWORK

 In data accusation, data storage, and

information delivery areas, the architectural
components serve specific purpose.
 Some of the architectural components
governs the flow of data from beginning to
end
 Some components perform standardizing,
summarizing, and cleansing of data

Data Warehousing/Mining 13
 ARCHITECTURAL FRAMEWORK

 At the data source level

– Data staging architectural component governs the
transformation, cleansing, and integration of data

 In DWH repository
– Meta data architectural component is also a storage
mechanism to contain data about data at every point of the
flow of data from beginning to end

 At user end
– The information delivery architectural component includes
dependent data marts, special multi dimensional databases,
and full range of query and reports facilities

Data Warehousing/Mining 14
 ARCHITECTURAL FRAMEWORK
 The management and control module
– It controls the entire DWH environment
– It has two major functions
 Constantly monitor all on going operations
 Recover from problems when things go wrong
– It manages and controls the data accusation functions
– Ensures that the extracts and transformations are carried out
correctly and in timely manners
– Manages baking up significant parts of DWH and
recovering from failure
– Monitoring the growth and periodically archiving data from
data warehouse
– Governs data security
– Provides authorized access to DWH
– Interfaces end user information delivery component to
ensure that information delivery is carried out properly
Data Warehousing/Mining 15
Data Warehousing/Mining 16
 Information Delivery Component

 Information is required for the wide

community
– Novice users
– Casual users
– Business analyst
– Power users

Data Warehousing/Mining 17
 Thank You Very Much

Data Warehousing/Mining 18