Data Warehouse Architecture - Evolving Application Domains of Data Warehousing and Mining - page 7

Table 1. Building blocks of data warehouse architecture

Basic architecture element Description

Operational system A system that captures the transactions of a line of business

Data staging area A storage area and a set of ETL and data quality processes

ETL operation An operation that does data extraction, transformation, or loading

Data mart A data silo with a subset of the whole enterprise data

Enterprise data warehouse A data warehouse with an integrated data model that keeps the whole enterprise data

Data access tools Tools, applications and user usage that create different BI output by accessing data in the data marts

Operational data store A repository of most recent operational data and accessed by special applications or users for fast

reporting and analytics

ture, the data staging area transforms the data
and applies data quality processes before the
data from operational systems is changed into
an integrated data model in the enterprise data
warehouse. The enterprise data warehouse nor-
mally utilizes a relational data model to keep the
different business data in the whole enterprise
in an integrated manner. The enterprise data
warehouse either feeds data to data marts which

are built upon star-schema models or directly
provide data to data access tools. The metadata
repository is a special tool that integrates meta-
data from different layers, components of the
architecture in a single place in order to provide
a single, unified view of all the metadata in the
whole architecture.
The prototype architecture can fulfill the 5
requirements by the following practices. First, to

Figure 4. Prototype data warehouse architecture