Scribd
Upload
242 views

Jeffrey A. Hoffer, Mary B. Prescott, Fred R. Mcfadden: Modern Database Management 10 Edition

This document discusses different architectures for data warehousing and data marts. It describes a generic two-level architecture with periodic extraction of data from operational systems to a centralized data warehouse. Independent data marts involve separate extraction and loading for each mart, while dependent data marts load from a single enterprise data warehouse. An operational data store can provide current operational data for decision making and act as a staging area for loading the enterprise data warehouse.

Uploaded by

Anees Malik
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
242 views

Jeffrey A. Hoffer, Mary B. Prescott, Fred R. Mcfadden: Modern Database Management 10 Edition

This document discusses different architectures for data warehousing and data marts. It describes a generic two-level architecture with periodic extraction of data from operational systems to a centralized data warehouse. Independent data marts involve separate extraction and loading for each mart, while dependent data marts load from a single enterprise data warehouse. An operational data store can provide current operational data for decision making and act as a staging area for loading the enterprise data warehouse.

Uploaded by

Anees Malik
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 13

Modern Database Management

10th Edition
Jeffrey A. Hoffer, Mary B. Prescott,
Fred R. McFadden

1
 Requirement for Integrated, company-
wide view of high-quality information
(from different databases)
 Separation of operational and
informational systems and data (for
improved performance)
◦ Operational systems is used to run the
business in real time, based on current data.
Also called system of record
◦ Information systems are designed to support
decision making based on historical data for
complex queries or data mining applications.

2
 Why organizations need to bring data from various
system of records?
 Answer: Speed and flexibility in decision making…

3
Data Mart: A data warehouse that is limited in
scope

4
 Generic Two-Level Architecture
 Independent Data Mart
 Dependent Data Mart and Operational Data
Store
 Logical Data Mart and @ctive Warehouse
 Three-Layer architecture

All involve some form of extraction, transformation and loading (ETL)

5
Figure 09-2: Generic two-level architecture (4 steps)

L
One,
company-
wide
T warehouse

Periodic extraction  data is not completely current in warehouse

6
Figure 09-3: Independent data mart
Data marts:
Mini-warehouses, limited in scope

T
E

Separate ETL for each Data access complexity


independent data mart due to multiple data marts

7
 Why created?
◦ Short-term business objectives
◦ Lower-cost to implement
◦ Organizationally and politically easier to implement
◦ Due to size limitation of the adopted technology
 Benefits
◦ We may work in series of small projects to prove
the concept of data warehousing
◦ Start getting benefits of data warehousing earlier
before waiting for all data to be centralized

8
1. Costly redundant data and processing due to
separate ETL process
2. Data marts may not be consistent with each other
3. Limited analysis (difficulty in making joins across
different platforms of separate data marts)
4. High cost of scaling for new data marts due to
expected changes in the technology over the time
5. High cost of making separate data marts consistent
with each other

9
Figure 09-4: ODS provides option for
Dependent data mart with operational data store obtaining current data

T
E Simpler data access
Single ETL for
enterprise data warehouse Dependent data marts
(EDW) loaded from EDW

10
 1st and 2nd limitation is addresses by loading the
dependent data marts from an enterprise data ware
(EDW) house which is single control point / version of
truth.
 Redundancy across DDM is planned and consistent
 Often called Hub and Spoke approach.
 The 3rd limitation is addressed by providing updatable
and current-valued operational data store (ODS) for
decision support processing to serve op users.
 Contrary to EDW, the ODS doesn't contain history data.
 ODS may be fed from ERP database and serves as
staging area from loading data to EDW.

11
 This is practical for only
◦ Moderate size data warehouse or
◦ When using high-performance DW technology,
such as Teradata (NCR)

12
ODS and data warehouse
are one and the same

T
E
Near real-time ETL for Data marts are NOT separate databases,
@active Data Warehouse but logical views of the data warehouse
 Easier to create new data marts
13

You might also like