Data warehouse

(IS 422)

Lecture 3
Main Data warehouse types

Dr. Wael Abbas

2023 - 2024
 Data warehousing
autonomous datawarehouse

• ADWH is a technique for collecting and managing data from varied sources
to provide meaningful business insights. It is a blend of technologies and
components which aids the strategic use of data. (Inmon Bill)

• Whereas data warehouse is a repository of data, data warehousing is

literally the entire process.

• Data warehousing is a discipline that results in applications that provide

decision support capability, allows ready access to business information, and
creates business insight.
 Data warehousing main types
• The three main types of data warehouses are :

– Data marts (DMs)

– operational data stores (ODS)

– enterprise data warehouses (EDW)

 Data Mart
• Whereas a data warehouse combines databases across an entire enterprise, a
data mart (DM) is usually smaller and focuses on a particular subject or
department.

• A DM is a subset of a data warehouse, typically consisting of a single

subject area (e.g., marketing, operations).

• A DM can be either dependent or independent.

 Data Mart
• A dependent data mart is a subset that is created directly from the data
warehouse. It has the advantages of using a consistent data model and
providing quality data.

• Dependent DMs support the concept of a single enterprise-wide data model,

but the data warehouse must be constructed first.

• A dependent DM ensures that the end user is viewing the same version of
the data that is accessed by all other data warehouse users. The high cost of
data warehouses limits their use to large companies.
 Data Mart
• Many firms use a lower-cost, scaled-down version of a data warehouse
referred to as an independent DM subset.

• independent data mart is a small warehouse designed for a strategic

business unit or a department, but its source is not an EDW.
Data warehouse
 operational data stores (ODS)
• Operational data store (ODS) provides a fairly recent form of customer
information file.
• This type of database is often used as an interim staging area for a data
warehouse.
• Unlike the static contents of a data warehouse, the contents of an ODS are
updated throughout the course of business operations.
• An ODS is used for short-term decisions involving mission-critical
applications rather than for the medium- and long-term decisions associated
with an EDW.
 operational data stores (ODS)
• An ODS is similar to short-term memory in that it stores only very recent
information. In comparison, a data warehouse is like long-term memory
because it stores permanent information.
• An ODS consolidates data from multiple source systems and provides a
near–real-time, integrated view of volatile, current data.
• The exchange, transfer, and load (ETL) processes for an ODS are identical
to those for a data warehouse.
• ODS is based on change data capture(CDC).This approach used to
determine the data change and apply action based on this change.
 Enterprise data warehouses (EDW)
• An enterprise data warehouse (EDW) is a large-scale data warehouse that
is used across the enterprise for decision support.
• The large-scale nature of an EDW provides integration of data from many
sources into a standard format for effective BI and decision support
applications.
• EDWs are used to provide data for many types of decision support systems
(DSS), including customer relationship management (CRM), supply chain
management (SCM), business performance management (BPM), business
activity monitoring, product life cycle management, revenue management,
and sometimes even knowledge management systems.
 Dimensional modeling
• Like enterprise relationship (ER) modeling, dimensional modeling is a
logical design technique.
• Dimensional modeling is much better suited for business intelligence (BI)
applications and data warehousing (DW).
represent
• It depicts business processes throughout an enterprise and organizes that
data and its structure in a logical way.
• The purpose of dimensional modeling is to enable BI reporting, query, and
analysis.
• The key concepts in dimensional modeling are facts, dimensions, and
attributes. There are different types of facts, depending on whether they
can be added together.
 Dimensional modeling
• Dimensions can have different hierarchies, and have attributes that define
the who, what, where, and why of the dimensional model. The grain, or
level of granularity, is another key concept with dimensional modeling, as
it determines the level of detail.
• Facts, dimensions, and attributes can be organized in several ways, called
schemas. The choice of schema depends on variables such as the type of
reporting that the model needs to facilitate and the type of BI tool being
used.
• Building a dimensional model includes additional puzzle pieces such as
calendar and time dimensions, and more complicated pieces such as
degenerative dimensions and consolidated fact tables.
 Dimensional modeling
• Facts
– A fact is a measurement of a business activity, such as a business event
or transaction, and is generally numeric. Examples of facts are sales,
expenses, and inventory levels. Numeric measurements may include
counts, dollar amounts, percentages, or ratios.
– A fact is a collection of related data consisting of measures.
– In a data warehouse, facts are implemented in the core tables in which
all of the numeric data is stored.
– Facts can be aggregated or derived. For example, you can sum up the
total revenue or calculate the profitability of a set of sales transactions.
 Dimensional modeling
• Facts
– A fact is a measurement of Facts provide the measurements of how
well or how poorly the business is performing.
– A fact is also referred to as organizational performance measure.
– Fact contains redundancy.
– Ninety-percent of the data in a dimensional model is typically located
in the fact tables.
– The key dimensional modeling design concerns when working with the
data in fact tables are how to minimize and standardize it and make it
consistent.
 Dimensional modeling
• Facts
– Fact tables are composed of two types of columns: keys and measures.

– The first, the key column, consists of a group of foreign keys (FK) that
point to the primary keys of dimensional tables that are associated with
this fact table to enable business analysis.

– The relationships between fact tables and the dimensions are one-to-
many
 Dimensional modeling
• Facts
 Dimensional modeling
• Facts
 Dimensional modeling
• Facts
– The second type of column in a fact table is the actual measures of the
business activity such as the sales revenue and order quantity.
– Every measurement has a grain, which is the level of detail in the measurement
of an event such as a unit of measure, currency used, or ending daily balance of
an account.
– For example SalesQuantity, SalesAmount, ReturnAmount, ReturnQuantity,
DiscountAmount, DiscountQuantity, and TotalCost that apply to a customer for
a product purchased at a specific time.
– All of these measures are related to the business event (the sale) that the fact
represents and they have a level of granularity related to that event.
 Dimensional modeling
• Facts
.

Measures in fact tables

 Dimensional modeling
• Dimensions
– A dimension is an entity that establishes the business context for the
measures (facts) used by an enterprise.
– Dimensions define the who, what, where, and why of the dimensional
model, and group similar attributes into a category or subject area.
Examples of dimensions are product, geography, customers, employees,
and time.
– Whereas facts are numeric, dimensions are descriptive in nature
(although some of those descriptions, such as a product list price, may
be numeric).
 Dimensional modeling
• Dimensions
 Schema Types
• Schema types are :
1. Star schema
2. Snowflake schema

You choose which schema to use when building the dimensional model by
considering these questions:
1. What kind of analysis are you trying to perform on that data and how complex is
it?
2. What are the analytical requirements and restrictions?
3. How consistent is the data you want to query and analyze?
4. What BI tool do you plan to use? Although different tools may appear to show
the same type of data, results, and graphs, they can be very different under the
covers, and rely on a specific schema for the best results.
Note : most slides in this file produced from : Sharda, Ramesh, Dursun Delen, and Efraim
Turban. Business intelligence, analytics, and data science: a managerial perspective. pearson,
 Schema Types
Star schema
• The star schema (sometimes referenced as star join schema) is the most
commonly used and the simplest style of dimensional modeling.
• A star schema contains a central fact table surrounded by and connected to
several dimension tables.
• The fact table contains a large number of rows that correspond to observed
facts and external links (i.e., foreign keys).
• A fact table contains the descriptive attributes needed to perform decision
analysis and query reporting, and foreign keys are used to link to
dimension tables.
 Schema Types
Star schema
• The decision analysis attributes consist of performance measures,
operational metrics, aggregated measures (e.g., sales volumes, customer
retention rates, profit margins, production costs, scrap rate), and all the
other metrics needed to analyze the organization’s performance.

• In other words, the fact table primarily addresses what the data warehouse
supports for decision analysis.

• Surrounding the central fact tables (and linked via foreign keys) are
dimension tables.
 Schema Types
Star schema
• The dimension tables contain classification and aggregation information
about the central fact rows.

• Dimension tables contain attributes that describe the data contained within
the fact table; they address how data will be analyzed and summarized.

• Dimension tables have a one-to-many relationship with rows in the central

fact table.

• In querying, the dimensions are used to slice and dice the numerical values
in the fact table to address the requirements of an ad hoc information need.
 Schema Types
Star schema
Star Schema Main Characteristics
1. Simplicity: It is the simplest type of DWH schemas.

2. Query effectiveness: Because of simplicity, It needs less join to

query the data (It is optimized to query large dataset).

3. Data Redundancy and Large Table Size: Due to de-

normalization, it has a data redundancy, and the table size is huge.

4. Most used and widely supported.

Note : This slide from : https://gability.com/en/courses/big-data-in-depth/02-dwh/03-
architecture/10-data-modeling/04-schema-types/
 Schema Types
Star schema
Star Schema additional Characteristics
1. Dimensions represented by one one-dimension table.

2. The dimension table are not joined to each other

3. The fact table would contain key and measure.

4. Data integrity is not enforced due to the de-normalized structure.

Note : This slide from : https://gability.com/en/courses/big-data-in-depth/02-dwh/03-

architecture/10-data-modeling/04-schema-types/
 Schema Types
Star schema
• The star schema is designed to provide fast query-response time,
simplicity, and ease of maintenance for read-only database structure.
 Schema Types
Snowflake schema
• Closely related to the star schema, the snowflake schema is represented by
centralized fact tables (usually only one), which are connected to multiple
dimensions.
• It takes the star schema, with the facts surrounded by denormalized
dimensions, one step further by normalizing the hierarchies within a
particular dimension.
• Each level in the dimensional hierarchy becomes its own dimensional table
with parent keys created to link the hierarchical structure together.
• The fact table stores the foreign key to the lowest level of the dimensional
hierarchy.
 Schema Types
Snowflake schema
• Dimensions are normalized into multiple related tables, whereas the star
schema’s dimensions are denormalized, with each dimension being
represented by a single table.
Snowflake Schema Characteristics
1. Extension: Snowflake is an extension of the Star Schema.

2. Normalized: Dimension tables are normalized; this means every

dimension may expand into additional tables.
 Schema Types
Snowflake schema .
Snowflake Schema Characteristics
3. Disk Space Efficiency: Due to its normalization methodology, it uses less
desk space, which enhances the query as we scan less data size.

4. Complicated: Due to the normalization query needs to join more table in

some cases to get the data which reduces the performance.
 Schema Types
Snowflake schema .
 Schema Types
Star vs snowflake example
 Schema Types
Star
 Schema Types
Snowflake
 Schema Types
Star Snowflake
Dimension represented by one-table Dimension tables are expanded into
multi-tables
Fact table surrounded by dimension Fact table surrounded by Hierarchy of
tables dimension tables
Less join Requires many joins
Simple Design Very Complex Design
De-normalized Data structure Normalized Data Structure
High level of Data redundancy Very low-level data redundancy
Maintenance is difficult Maintenance is easier
Good for datamarts with simple Good for core to simplify (many:
relationships (1:1 or 1:many) many)
 Schema Types
Multi-fact star models
 Examples of dimensional models, whether star, snowflake, or
multidimensional, usually depict them with a single fact table
surrounded by dimensions. Although this is a popular illustration
and a terrific approach that simplifies explanations, the reality is
that data in an enterprise requires multiple facts.
 Examples of facts are sales, expenses, inventory, and the many
other business events associated with running an enterprise.
 For example , shows two facts: store sales and store inventory
(Tbl_Fact_Store_Sales and Tbl_Fact_ Store_Inventory).
 Schema Types
Multi-fact star models
 On the right side are three dimensions that both facts share: the item
dimension, the date dimension, and the buyer dimension. The shared
dimensions are referred to as conformed dimensions.
 On the left side of the example are the customer and store dimensions
that are not shared across the two facts. The Tbl_Dim_Customer is
related to Tbl_Fact_Store_Sales, whereas the Tbl_Dim_Store is related
to Tbl_Fact_Store_Inventory
 Schema Types
Multi-fact star models
 ENTITY RELATIONSHIP vs
DIMENSIONAL MODELING
 ER modeling, sometimes referred to as normalized modeling, is the standard
for transactional systems (also called operational or online transactional
processing (OLTP) systems).
 Dimensional modeling is the best practice for BI and OLAP systems. The
reason for this stems from how the two systems process and manipulate data.
 The data in operational systems(OLTP) is being constantly updated. The
main job of these workhorses is transaction throughput. They have to
maintain and update a large number of records, so they have to be able to
handle high amounts of consistent throughput with large volumes.
 Compare that with the BI and reporting systems, which are generally not
updating data. Rather, they are copying, extracting, transforming, and loading
from operational systems. They are not doing realtime updates like
operational systems.
 ENTITY RELATIONSHIP vs DIMENSIONAL
MODELING
 The focus is not on transactional throughput once the data is loaded. The
focus is on query performance, and gathering and aggregating large sets of
data. That’s a critical operation. Transactional systems update a small numbe
of data records quite often, whereas BI systems gather and aggregate large
sets of data records, but not so often—sometimes over large spans of time.
 The data from the BI system then get used for reports.
 The operational system and its accompanying normalized models have
minimal redundancy to support the high transaction throughput.
 in BI reporting, there is a lot of redundancy. After all, the warehouse and
data marts and whatever else you store on the reporting side are all copies of
data that originally was in the operational systems. The BI system depends
heavily on indexes and partitioning because it needs much more storage
space for reporting.
ENTITY RELATIONSHIP vs DIMENSIONAL
MODELING

non-voltile
updated