CH7: Using Data Flow Diagrams Data Store

§ A depository for data that allows examination, addition, and

Data Flow Diagrams retrieval of data
§ Graphically characterize data processes and flows in a § Named with a noun, describing the data
business system § Data stores are usually given a unique reference number,
§ Depict: such as D1, D2, D3
o System inputs § Represents a:
o Processes o Database
o Outputs o Computerized file
o Filing cabinet
Advantages of the Data Flow Approach
§ Freedom from committing to the technical implementation Steps in Developing Data Flow Diagrams
too early
§ Understanding of the interrelatedness of systems and
subsystems
§ Communicating current system knowledge to users
§ Analysis of the proposed system

Basic Symbols
§ A double square for an external entity
§ An arrow for movement of data from one point to another
§ A rectangle with rounded corners for the occurrence of a
transforming process
§ An open-ended rectangle for a data store

The Four Basic Symbols Used in Data Flow Diagrams, Their

Meanings, and Examples

Creating the Context Diagram

§ The highest level in a data flow diagram
Contains only one process, representing the entire system
§ The process is given the number 0
§ All external entities, as well as major data flows are shown

Basic Rules
External Entities § The data flow diagram must have one process
§ Represent another department, a business, a person, or a § Must not be any freestanding objects
machine § A process must have both an input and output data flow
§ A source or destination of data, outside the boundaries of the § A data store must be connected to at least one process
system § External entities should not be connected to one another
§ Should be named with a noun
Context Diagram
Data Flow
§ Shows movement of data from one point to another
§ Described with a noun
§ Arrowhead indicates the flow direction
§ Represents data about a person, place, or thing

Process
§ Denotes a change in or transformation of data
§ Represents work being performed in the system Drawing Diagram 0
§ Naming convention: § The explosion of the context diagram
o Assign the name of the whole system when naming § May include up to nine processes
a high level process § Each process is numbered
o To name a major subsystem attach the word § Major data stores and all external entities are included
subsystem to the name § Start with the data flow from an entity on the input side
o Use the form verb adjective noun for detailed § Work backward from an output data flow
processes § Examine the data flow to or from a data store
§ Analyze a well-defined process
§ Take note of any fuzzy areas
Note Greater Detail in Diagram 0 Checking the Diagrams for Errors
§ Forgetting to include a data flow or pointing an arrow in the
wrong direction

§ Connecting data stores and external entities directly to each

other

Data Flow Diagram Levels

§ Data flow diagrams are built in layers
§ The top level is the context level
§ Each process may explode to a lower level
§ The lower-level diagram number is the same as the parent
process number
§ Processes that do not create a child diagram are called
primitive
Typical Errors that Can Occur in a Data Flow Diagram (Payroll
Creating Child Diagrams Example)
§ Each process on diagram 0 may be exploded to create a child
diagram
§ A child diagram cannot produce output or receive input that
the parent process does not also produce or receive
§ The child process is given the same number as the parent
process
o Process 3 would explode to Diagram 3
§ Entities are usually not shown on the child diagrams below
Diagram 0
§ If the parent process has data flow connecting to a data store,
the child diagram may include the data store as well
§ When a process is not exploded, it is called a primitive
process

DiLerences between the Parent Diagram (above) and the Child

Diagram (below)

Logical and Physical Data Flow Diagrams

§ Logical
o Focuses on the business and how the business
operates
o Not concerned with how the system will be
constructed
o Describes the business events that take place and
the data required and produced by each event
§ Physical
o Shows how the system will be implemented
o Depicts the system

Features Common of Logical and Physical Data Flow Diagrams

Data Flow Diagrams Error Summary

§ Forgetting to include a data flow or pointing an arrow in the
wrong direction
§ Connecting data stores and external entities directly to each
other
§ Incorrectly labeling processes or data flow
§ Including more than nine processes on a data flow diagram
§ Omitting data flow
§ Creating unbalanced decomposition (or explosion) in child
diagrams
The Progression of Models from Logical to Physical CRUD Matrix
§ The acronym CRUD is often used for
o Create, Read, Update, Delete
§ These are the activities that must be present in a system for
each master file
§ A CRUD matrix is a tool to represent where each of these
processes occurs in a system

Logical Data Flow Diagram Example

Event Modeling and Data Flow Diagrams

§ An input flow from an external entity is sometimes called a
trigger because it starts the activities of a process
§ Events cause the system to do something and act as a trigger
to the system
§ An approach to creating physical data flow diagrams is to
create a data flow diagram fragment for each unique system
Physical Data Flow Diagram Example
event

Event Response Tables

§ An event table is used to create a data flow diagram by
analyzing each event and the data used and produced by the
event
§ Every row in an event table represents a data flow diagram
fragment and is used to create a single process on a data flow
diagram

Developing Logical Data Flow Diagrams An Event Response Table for an Internet Storefront
• Better communication with users
• More stable systems
• Better understanding of the business by analysts
• Flexibility and maintenance
• Elimination of redundancy and easier creation of the
physical model

Developing Physical Data Flow Diagrams

• Clarifying which processes are performed by humans
and which are automated
• Describing processes in more detail
• Sequencing processes that have to be done in a
particular order
• Identifying temporary data stores
• Specifying actual names of files and printouts Data Flow Diagrams for the First Three Rows of the Internet
• Adding controls to ensure the processes are done Storefront Event Response Table
properly

Physical Data Flow Diagrams Contain Many Items Not Found in

Logical Data Flow Diagrams
Use Cases and Data Flow Diagrams
§ Each use case defines one activity and its trigger, input, and
output
§ Allows the analyst to work with users to understand the
nature of the processes and activities and then create a single
data flow diagram fragment

Partitioning Data Flow Diagrams

§ Partitioning is the process of examining a data flow diagram
and determining how it should be divided into collections of
manual procedures and computer programs
§ A dashed line is drawn around a process or group of
processes that should be placed in a single computer
program

Reasons for Partitioning

• DiXerent user groups
• Timing
• Similar tasks
• EXiciency
• Consistency of data
• Security

Partitioning Websites
• Improves the way humans use the site
• Improves speed of processing
• Ease of maintaining the site
• Keep the transaction secure

Communicating Using Data Flow Diagrams

§ Use unexploded data flow diagrams early when ascertaining
information requirements
§ Meaningful labels for all data components

Summary
§ Data flow diagrams
o Structured analysis and design tools that allow the
analyst to comprehend the system and subsystems
visually as a set of interrelated data flows
§ DFD symbols
o Rounded rectangle
o Double square
o An arrow
o Open ended rectangle
§ Creating the logical DFD
o Context level data flow diagram
o Level 0 logical data flow diagram
o Child diagrams
§ Creating the physical DFD
o Create from the logical data flow diagram
o Partitioned to facilitate programming
§ Partitioning data flow diagrams
o Whether processes are performed by diXerent user
groups
o Processes execute at the same time
o Processes perform similar tasks
o Batch processes can be combined for eXiciency of
data
o Processes may be partitioned into diXerent
programs for security reasons