Tài liệu Module 4: Deriving a Logical Data Design doc

Module 4: Deriving a Logical Data
Design


THIS PAGE LEFT INTENTIONALLY BLANK




Module 1: Course
Overview
Module 4: Deriving a
Logical Data Design
Module 5: Normalizing the
Logical Data Design
Module 6: Deriving a
Physical Data Design
Module 7:
Implementing Data
Integrity
Module 2: Solution
Design Processes
Module 3: Using a
Conceptual Design for Data
Requirements
Module 8: Designing
Data Services
Module 9: Data Storage
Considerations
Designing Data

Services and
Data Models
Logical Data
Design
Entities and
Attributes
Activity 4.1: Deriving
Entities and Attributes from
Data Requirements
Data
Relationships
Entity/Relationship
Modeling
Activity 4.2: Creating a
Logical Data Model
Module 4: Deriving a
Logical Data Design

Module 4: Deriving a Logical Data Design 67



!
!!
! Overview
"
Logical Data Design
"
Entities and Attributes
"

Data Relationships
"
Entity/Relationship Modeling
"
Review
In this module...
In this module...


At the end of this module, you will be able to:
"
Analyze data requirements to determine data entities and attributes.
"
Analyze data entities and attributes to determine their relationships.
"
Determine the cardinality and existence characteristics of a relationship.
"
Create an entity/relationship diagram.

Slide Objective
To provide an overview of
the module topics and
objectives.
Lead-in
In this module, you will learn
about logical data design
and entity/relationship
modeling.
68 Module 4: Deriving a Logical Data Design

!
!!
! Logical Data Design
"
The Need for a Logical Data Design
"
Deriving Logical Data Services
In this section...
In this section...


In this section, you will learn about logical data design. During this stage of the
design process, you will develop a data structure and formalize the solution’s
conceptual design.
Slide Objective
To introduce the concept of
logical data design.
Lead-in
This section discusses the
logical data design of a
solution.
Module 4: Deriving a Logical Data Design 69



The Need for a Logical Data Design
"
A logical data design:

$
Translates conceptual data needs into real data
requirements
$
Helps organize thinking about the data
$
Focuses on the semantic details of data relationships
$
Provides a mechanism for communication


The logical data design is the middle stage of the natural progression from
conceptual design to physical design. The process of deriving a logical data
design from the conceptual data design accomplishes the following:
"
Translates conceptual data needs into real data requirements.
You must turn the conceptual data needs identified during conceptual design
into actual entities and relationships that will define how the data interacts.
"
Helps organize your thinking about the data.
You must ensure that the data elements identified in the conceptual design
stage accurately depict the desired data for the business solution.
"
Focuses on the semantic details of the data relationships.
You are forced to delve deeper into the solution’s data requirements and
look at how data interacts within both the existing system and the proposed
solution.
"
Provides a mechanism for communication.
The logical design helps to document the data design for all interested

parties to read and understand.

Slide Objective
To discuss the reasons why
a logical data design is
necessary.

70 Module 4: Deriving a Logical Data Design



Deriving Logical Data Services
"
Organize data requirements into entities
"
Identify data services
"
Identify logical groups of services
"
Identify the relationship between groups and entities
"
Refine the design


Some of the questions that must be addressed during logical data design
concern the data services, which determine how the data will be managed and
manipulated. Development of a logical design provides a road map of these
services.
Organize data requirements into entities
After the data requirements for a set of use cases have been collected, you need

to organize these data requirements into similar groups of information. The
groupings of requirements are referred to as entities. The entities allow you to
track all of the required data on a given subject.
Identify data services
It is important to identify any services that will act upon data.
For example, consider the timesheet from the Ferguson and Bardell, Inc. case
study. Many steps must be taken to complete the timesheet during its life span.
Many employees will edit the timesheet or read its information. An automated
process may extract timesheet information into another database for reporting
purposes.
After determining how a particular entity will be used, you determine whether a
service should exist that will act as the interface to that entity.
Identify logical groups of services
"
While developing the data services, you might find it beneficial to group
services that act upon common data or upon common processes. These
groups of services can help bring consistency to the solution, whether they
concern a process that manipulates an employee object or a set of objects
that deal with bank accounts.
Slide Objective
To introduce the logical data
services.
Lead-in
When developing a solution,
you must address questions
regarding the data services.
Module 4: Deriving a Logical Data Design 71

For example, consider the consultant, consultant manager, and
administrative assistant from the Ferguson and Bardell, Inc. case study. All
of them have a need to interact, to some degree, with time and billing
information that is stored in the system. One data service might handle the
submittal of timesheets from the consultants. A second data service might
handle the processing of timesheet data into billable data for the
administrative assistants. The fact that all users of the Ferguson and Bardell,
Inc. system need to work with timesheet data does not necessarily mean that
one all-encompassing data service should handle everything that needs to be
done to the timesheet as a single category of data.
"
After the logical data services are grouped by functionality, you can design
data services that handle the logical functionality needed by the solution.

Identify the relationship between groups and entities
"
The data that the services act upon typically does not exist in a vacuum and
most real-life situations require that some type of relationship exist between
data. Relationships between the services and data should be clearly defined.
Identifying the relationships provides greater logical design definition and
enhances the functionality of the logical design.
"
At the logical level, it is important to know what relationships exist between
data. Later, these relationships will become part of the logical data model,
which is a representation of the data; the users will work with the logical
data model on a daily basis.
"
By identifying all the possible relationships, the development team can
better understand the complete picture.

Refine the design
"
After organizing and analyzing entities, you will have to revisit the initial
design and make refinements where necessary. This reiteration through the
design is a necessary process that will yield a more thorough and complete
design.
72 Module 4: Deriving a Logical Data Design



!
!!
!

Entities and Attributes
"
Overview of Entities and Attributes
"
Deriving Entities
"
Deriving Attributes
"
Activity 4.1: Deriving Entities and Attributes from Data
Requirements
In this section...
In this section...


In this section, you will learn about logical entities and their attributes as
represented in a logical model.

Slide Objective
To introduce entities and
attributes.
Lead-in
The data services are
represented in the logical
design by entities and
attributes.
Module 4: Deriving a Logical Data Design 73



Overview of Entities and Attributes
"
Entities and their corresponding attributes are the
building blocks of relational design
"
An entity is anything about which data will be collected
and stored
$
An entity is a noun
$
Instances of entities are stored in the rows of a table
"
An attribute is a characteristic of an instance of an
entity
$
Attributes are the adjectives
$
Attributes are stored in the columns of a table

When you move into the logical stage of data design, one of your first tasks is
to formulate entities from data requirements and other related information.
An entity can be thought of as any person, place, thing, or concept that defines
data or about which data will be collected and stored.
An attribute is a characteristic that further defines and describes the properties
of an instance of an entity. An entity can have multiple attributes.
Think of entities as nouns and attributes as adjectives.
Entities and attributes identified during logical design will be used later to help
model the physical design.
Slide Objective
To introduce the
fundamental concepts of
entities and attributes.
Lead-in
Entities and attributes form
the basis for logical design.
74 Module 4: Deriving a Logical Data Design



Deriving Entities
"
Entities represent real-world objects about which
information will be stored
"
When deriving entities, look for nouns or noun phrases
during analysis
"

Rows in database tables
"
Common examples
$
People
$
Books


When deriving entities, keep in mind that entities are:
"
Objects about which information will be stored.
The entities are the objects about which you wish to retain information.
Some examples of entities are Employees, Clients, and Timesheets.
"
The starting point for the logical data design.
The identification of these entities is the first step in the design of a
database.
"
The equivalent of rows in database tables.
An instance of an entity will correspond to a row of a table.

During requirements analysis, the following techniques will help you derive
entities:
"
Find use cases from gathered information.
"
Look, or listen, for nouns or noun phrases within the use cases. Nouns
describe people, places, things, or concepts.

Consider the following example:
“Consultants enter their time into a timesheet on a weekly basis. The timesheet
is then forwarded to the administrative assistant, who keys the time into the
invoicing program. The administrative assistant then sends invoices to the
customers based on the time reported.”
The italicized words are possible candidates for entities. After you have
identified the entities, it is sometimes useful to write a short, one-sentence
entity definition that will help determine relationships at a later time.
Slide Objective
To introduce methods for
extracting entities from data
requirements.
Lead-in
When deriving the entities of
a solution, keep the
following factors in mind.
Delivery Tip
Point out that entities can
represent anything, from
people to books to hats.
Module 4: Deriving a Logical Data Design 75



Deriving Attributes
"
Descriptive information about an entity
"
Attached to entity that they most closely describe
"

Columns in database tables
"
Example: Attributes of a car
$
Color
$
Make
$
Model
$
Year


After identifying the entities, you must determine which of their attributes you
want to capture. Attributes have the following characteristics:
"
Describe an entity.
Attributes describe the solution’s entities. For example, the attributes of a
car may be its color, make, model, and year of production.
"
Should be attached to the entity that they most closely describe.
An attribute can exist only when attached to an entity. For example, the
attribute of color does not describe anything tangible unless the color is
applied to an object.
"
Define the columns in database tables.
When the physical design is implemented, the attributes become the
columns in the database tables.

There is no specific method for identifying attributes. Iteratively brainstorming

about the characteristics of an entity can be sufficient. If determining attributes
is difficult, you should check whether the definitions of the entities are
sufficiently specific.
Slide Objective
To introduce attributes.
Lead-in
After identifying the entities,
you must determine which
of their attributes you want
to capture.