Classes 19, 20, 22, & 23 Classical Analysis September 24, 26, 29, & October 1, 2014

Software Engineering CSCI 4490
Classes 19, 20, 22, & 23

Classical Analysis
September 24, 26, 29, &

October 1, 2014
Outline
Products of Classical Analysis Workflow

Specification Document
Software Project Management Plan
Software Specification Methods

Informal
Semi-Formal
Structured Systems Analysis
Entity-Relationship Modeling (ER Diagrams)
Formal
Finite State Machines (FSMs)
Petri Nets
Z
2
* The Specification Document (The What)
Detailed description of what system will do
Faults in specification lead to faults in final software, so
Considerations:
feasibility - can s/w be built as specified?
Q: How to check specification correctness?
3
The Classical Analysis Workflow (Schach Ch 11)
Specification document must be

Informal enough for client to understand, but
Formal enough for developers, and be contractually binding.
Free of omissions, contradictions, ambiguities
Acceptance Criteria determinable as part of Specification

Can even begin (some) test case development based on
Requirements Document without yet having any code!
If product passes tests, deemed to satisfy specifications
4
Software Project Management Plan (The roadmap)
Once the client has signed off the specifications,

detailed planning and estimating begins
We draw up the software project management plan,

including
What?
How much?
When?
What is the earliest possible time for the SPMP?
The McGraw-Hill Companies, 2011

5
*Formality Versus Informality
Informal method: Use natural language like English.

Is this a good way to specify software? Why/Why Not?
6
Formality Versus Informality
Informal method: Use natural language like English.
Semiformal methods Formal methods

Structured Systems Analysis Finite State Machines
Entity-Relationship (FSMs)
Modeling (ER Diagrams) Petri Nets
Z
7
Structured Systems Analysis
Three popular graphical specification methods of 70s

DeMarco
Gane and Sarsen
Yourdon and Constantine
All are equivalent
All are equally good

8
*Data Flow Diagram- Foundation of Gane and
Sarsens Structured System Analysis
Graphical Notation used to Describe how data flows between

Processes in a System:
Use Notation That Represents Functional Processing, Data
Stores and Data Movements (flows) Between Sequences of
Functional Units.
Focus is on What'
Happens, Not 'How'
It Happens.
Development
Proceeds Stepwise
Start with Context
Level diagram, and
then refine.
9
Sallys Software An Example Using DFDs
Sallys Software Store buys software from various

suppliers and sells it to the public:
Popular software packages are kept in stock, but the
rest must be ordered as required.
Institutions and corporations are given credit facilities,
as are some members of the public.
Sallys Software Store is doing well, with a monthly
turnover of 300 packages at an average retail cost of
$250 each.
Sally hires you to help computerize her business.

10
Sallys Software Shop Mini Case Study (cont.)
The fundamental issue

What is Sallys objective in computerizing her
business?

11
*Sallys Software Shop Mini Case Study (cont.)
The danger of many standard approaches:
Gane and Sarsens method

Nine-step method
Stepwise refinement is used in many steps

12
How to Perform Structured Systems Analysis
1. Draw the data flow diagram

2. Decide what sections to computerize and how (batch
or online)
3. Determine the details of the data flows
4. Define the logic of the processes
5. Define the data stores
6. Define the physical resources
7. Determine the input-output specifications
8. Perform the sizing
9. Determine the hardware requirements

13
*Step 1. Draw the DFD
First refinement
Infinite number of possible interpretations

14
*Step 1. (cont.)
Second refinement
Verify that order is valid
Differentiate between items in/out of stock

15
*Step 1. (cont.)
Portion of third refinement

Add invoice/payment details

16
Step 1. (cont.)
The final DFD is larger

But it is easily understood by the client
When dealing with larger DFDs

Set up a hierarchy of DFDs
A box becomes a DFD at a lower level

17
~ICE/Homework: Using DFDs
For the Course Project, use a DFD to model the

business process of the project proposed by your
team.
18
Step 2. Decide What Parts to Computerize and How
Depends on how much client is prepared to spend

Cost/benefit analysis
Determine solution strategy (How)
General approach to building the product
Find strategies without worrying about constraints
Keep written record of all discarded strategies, and why discarded
Strategy dependent on volume, level of control

Large volumes, tight controls
Small volumes, in-house microcomputer

19
*Step 3. Determine the Details of the Data Flows
Determine the data items for each data flow
Refine each flow stepwise

Example;
We need a data dictionary for larger products

20
Sample Data Dictionary Entries
Figure 11.5
21
Step 4. Define the Logic of the Processes
For process create_invoice Sally has practice of

providing a discount to educational institutions
Sally explains the details of the discount she provides:
10% on up to 4 packages
15% on 5 or more

22
*Step 4 . Define the Logic of the Processes (cont.)
Translate this into a decision tree
What is the advantage of a decision tree?
23
* Step 5. Define the Data Stores
Define the exact contents and representation (format)

Specify the data structure(s) and data types to be used
Specify where immediate access is required

Data immediate-access diagram (DIAD)
Figure 11.8
24
* Step 6. Define the Physical Resources
For each file, specify:

25
Step 7. Determine Input/Output Specifications
Specify:

26
Step 8. Determine the Sizing
Obtain the numerical data needed to determine the

hardware requirements in Step 9

27
Step 9. Determine the Hardware Requirements
Mass storage requirements
Mass storage for back-up
Input needs
Output devices
Is the existing hardware adequate?

If not, recommend whether to buy or lease additional
hardware
28
Entity-Relationship Diagrams (Semi-Formal)
For specifying databases divide your database in two logical parts,

entities (e.g. "customer") and relationships ("buys", "pays for").
Author Supplier
1
m
writes
is supplied by p
Project
for use in
n
Autobiog raphy n
n n
Part
reads owns
1 n
1 1
Reader
consists of
Also for OO analysis in determining Objects a system will involve.
29
Data Modeling Notation
(b) Crows foot version (ERWin)
30
Data Modeling Notation: ERwin
31
Data Modeling Notation: N:M and O-M
Note that:
(1) ERwin cannot indicate
true minimum
cardinalities on N:M
relationships
(2) Visio introduces the
intersection table instead
of using a true N:M
model
32
*The Role of Data Dictionaries
Generally need further description of named entities

to make model understandable. Collect all
descriptions in a data dictionary:
Advantages:
Q? What data dictionary entries are needed to support your

class project ER model?
33
~ICE/Homework: Using ER Diagrams
For the Course Project, identify a minimum of four

data elements projected to be used by the project
proposed by your team (a minimum of two of which
shall be related in some manner).
Use an ER diagram to model the relationships
exhibited among the data elements selected above.
34
*Finite State Machines (Formal Method)
Formal Model of Different States of the System.

Useful for Software Specification. Formal here means
Mathematical Rigor used to ensure a proper FSM.
A FSM Has 5 Component Parts:
35
* Finite State Machine Example
A safe has a combination lock that can be in one of

three positions, labeled 1, 2, and 3.
The dial can be turned left or right (L or R). Six
possible dial movements, 1L, 1R, 2L, 2R, 3L, and 3R.
The combination to the safe is 1L, 3R, 2L; any other dial
movement will cause the alarm to go off

36
~*Finite State Machines (cont.)
States {Locked, A, B, Safe Unlocked, Sound Alarm}

Set of inputs {1L, 1R, 2L, 2R, 3L, 3R}
Initial state is Locked
Set of final states {Unlocked, Sound Alarm}
Transition function (Transition Table) takes current state and an
input event and returns new set of output events and next state:
Transition Table for Combo Lock
Current
State Locked A B Sound Alarm Safe
Dial Unlocked
Movement
1L A
1R Sound Alarm
2L
2R
3L
3R
37
*Analysis of FSMs
Power of FSMs - serves as an abstraction of the behavior

Advantages
Drawbacks?
38
*Where are FSMs used to Specify Software?
Commercial products
System software
Real-time systems
Statecharts are a real-time extension of FSMs
CASE tools typically incorporate FSM tool: Statemate,
Rhapsody
39
~Example: Apartment Automatic Garage Door
Description:
The door of the basement garage in an apartment building is opened
from the outside by means of a magnetic card. From the inside it is
opened by a floor sensor tripped by an exiting car. There is also a
manual unit with open, close, and stop buttons inside the garage.
Sensors in the door frame are tripped when the door reaches the fully
closed and fully opened positions. An optical sensor mounted on
the door frame ensures that the doorway is free from obstacles,
stopping the door from closing if the light beam is broken. When the
door has been either fully opened or stopped for 1 minute, it
automatically starts moving down, provided there is no obstacle.
Homework: Draw a Finite State Machine that depicts

the operation of the Apartment Automatic Garage
Door.
40
Petri Nets
A major difficulty with specifying real-time systems

is timing
Synchronization problems
Race conditions
Deadlock
Often a consequence of poor specifications

41
Petri Nets (cont.)
Petri nets
A powerful technique for specifying systems that have
potential problems with interrelations
A Petri net consists of four parts:

42
Petri Nets (cont.)
Set of places P is
{p1,p2, p3, p4}
Set of transitions T
is {t1, t2}
Input functions:
I(t1) = {p2, p4}
I(t2) = {p2}
Output functions:
Figure 11.18
O(t1) = {p1}
O(t2) = {p3, p3}

43
Petri Nets (cont.)
A marking of a Petri net is an assignment of tokens to

that Petri net
Figure 11.19
Four tokens: one in p1, two in p2, none in p3, and one in
p4
Represented by the vector (1,2,0,1)
44
*Petri Nets (cont.)
A transition is enabled if each of its input places has

as many tokens in it as there are arcs from the place
to that transition. Which transitions are enabled
(ready to fire)?

45
Petri Nets (cont.)
Petri nets are indeterminate

Suppose t1 fires
What is the resulting marking?
Figure 11.20

46
Petri Nets (cont.)
Now only t2 is enabled
Figure 11.21
What is the marking after it fires?

47
Petri Nets (cont.)
Inhibitor arcs
An inhibitor arc is marked by a small circle, not an
arrowhead
Figure 11.22
Transition t1 is enabled

48
Petri Nets (cont.)
Petri nets can also be used for design
Petri nets possess the expressive power necessary for

specifying synchronization aspects of real-time
systems
49
Z
Z (pronounced zed) is a formal specification

language
A Z specification consists of four sections:
1. Given sets, data types, and constants
2. State definition
3. Initial state
4. Operations

50
1. Given sets
Given sets are sets that need not be defined in detail
Names appear in brackets
For the elevator problem we need the set of all buttons
The specification therefore begins

[Button]

51
2. State Definition
Z specification consists of a number of schemata

A schema consists of a group of variable declarations,
plus
A list of predicates that constrain the values of
variables
Figure 11.25

52
Z: The Elevator Problem Case Study
In this problem there are four subsets of Button

The floor buttons
The elevator buttons
buttons (the set of all buttons in the elevator problem)
pushed (the set of buttons that have been pushed)

53
Schema Button_State
Figure 11.26

54
3. Initial State
The state when the system is first turned on
Button_Init ^= [Button_State' | pushed' = ]
(The caret ^ should be on top of the first equals sign =.

Unfortunately, this is hard to type in PowerPoint!)

55
4. Operations
Figure 11.27

56
Z: The Elevator Problem Case Study (cont.)
Figure 11.28

57
Analysis of Z
Z is the most widely used formal specification

language
It has been used to specify

CICS IBM Transaction Processing System (part)
An oscilloscope
A CASE tool
Many large-scale projects (especially in Europe)

58
Analysis of Z (cont.)
Difficulties in using Z
The large and complex set of symbols
Training in mathematics is needed

59
Analysis of Z (cont.)
Reasons for the great success of Z

It is easy to find faults in Z specifications
The specifier must be extremely precise
We can prove correctness (we do not have to)
Only high-school math needed to read Z
Z decreases development time
A translation of a Z specification into English (or

another natural language) is clearer than an informal
specification

60
Analysis Testing
The analysis artifacts should be checked by means of

a review
Representatives of the client and analysis team must be
present
Walkthroughs and inspections particularly effective
The SPMP must be similarly checked

Pay special attention to the cost and duration estimates

61
Specification Document Contents (Classical)
1. Introduction (Overview/Problem Statement)

2. Glossary (Data Dictionary)
3. Operating Environment (Environment in which system will run)
4. Interfaces (GUIs, Databases/Datastores, DFDs, FSMs)
5. System Operation (Major ops, internal ops, DFSs, FSMs, ER)
6. Information (Major units of data that system uses, ER diagrams)
7. Performance (Performance parameters to which system must
conform, min/max #users, timing constraints)
8. Parallelism (Portions of system that execute in parallel)
9. Concurrent Engineering (Portions of the system that can be
developed in parallel)
10. Security (What access must be controlled, passwords)
62

Classes 19, 20, 22, & 23 Classical Analysis September 24, 26, 29, & October 1, 2014

Cargado por

Información del documento

Título original

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

Classes 19, 20, 22, & 23 Classical Analysis September 24, 26, 29, & October 1, 2014

Cargado por

Copyright:

Formatos disponibles

Software Engineering CSCI 4490

Classes 19, 20, 22, & 23

September 24, 26, 29, &

Products of Classical Analysis Workflow

Software Specification Methods

Detailed description of what system will do

Faults in specification lead to faults in final software, so

Specification document must be

Acceptance Criteria determinable as part of Specification

Once the client has signed off the specifications,

We draw up the software project management plan,

What is the earliest possible time for the SPMP?

The McGraw-Hill Companies, 2011

Informal method: Use natural language like English.

Informal method: Use natural language like English.

Semiformal methods Formal methods

Three popular graphical specification methods of 70s

Yourdon and Constantine

All are equivalent

All are equally good

The McGraw-Hill Companies, 2011

Graphical Notation used to Describe how data flows between

Sallys Software Store buys software from various

The McGraw-Hill Companies, 2011

The fundamental issue

The McGraw-Hill Companies, 2011

The danger of many standard approaches:

Gane and Sarsens method

The McGraw-Hill Companies, 2011

1. Draw the data flow diagram

The McGraw-Hill Companies, 2011

The McGraw-Hill Companies, 2011

The McGraw-Hill Companies, 2011

Portion of third refinement

The McGraw-Hill Companies, 2011

The final DFD is larger

When dealing with larger DFDs

The McGraw-Hill Companies, 2011

For the Course Project, use a DFD to model the

Depends on how much client is prepared to spend

Find strategies without worrying about constraints

Keep written record of all discarded strategies, and why discarded

Strategy dependent on volume, level of control

Small volumes, in-house microcomputer

Determine the data items for each data flow

Refine each flow stepwise

We need a data dictionary for larger products

The McGraw-Hill Companies, 2011

For process create_invoice Sally has practice of

The McGraw-Hill Companies, 2011

Translate this into a decision tree

What is the advantage of a decision tree?

Define the exact contents and representation (format)

Specify where immediate access is required

For each file, specify:

The McGraw-Hill Companies, 2011

The McGraw-Hill Companies, 2011

Obtain the numerical data needed to determine the

The McGraw-Hill Companies, 2011

Mass storage requirements

Mass storage for back-up

Is the existing hardware adequate?

For specifying databases divide your database in two logical parts,