0% found this document useful (0 votes)
5 views26 pages

Chapter 11 TP11

Uploaded by

Mer Aba
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
5 views26 pages

Chapter 11 TP11

Uploaded by

Mer Aba
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 26

Chapter 11

Case Study
Review
 Object-oriented design is based on data abstraction and every basic system
component is a module supporting data abstraction.
 A class defines a possible set of objects.
 An object-oriented life-cycle approach applies object-oriented principles to
every phase of the development life cycle.
 A common process framework (CPF) characterizes an organization’s approach
to software development and maintenance.
 The component process assembly process model is commonly used as an OO
process model.
 Object-oriented metrics are different due to factors such as localization,
encapsulation, information hiding, inheritance, and object abstraction.
 Metrics used for object-oriented projects are:
 Number of scenario scripts

 Number of key classes

 Number of support classes

 Average number of support classes


Software Engineering Principles / Chapter 11 / 2 of 26
 Number of subsystems
Objective
 Study the implementation of configuration management in small
enterprises

Software Engineering Principles / Chapter 11 / 3 of 26


Business Motivation and
Objectives
 The Experiment

 Expected Impact and Experience

Software Engineering Principles / Chapter 11 / 4 of 26


The Experiment
 ICONMAN aims to implement lightweight configuration
management procedures for very small enterprises

 The procedures will be based on the current software


engineering practice of the partners

 Use of configuration management will address documentation


and error reporting

 The project will define and test a set of software metrics

Software Engineering Principles / Chapter 11 / 5 of 26


Expected Impact and
Experience
 Three companies will implement procedures for revision and
configuration management
 There will be procedures to handle all communication with the
customers that relates to product development
Business Sector Software Industry

Application Area Oil-reservoir simulation, statistical analysis,


actuarial software

Keywords Configuration Management, Software metrics

Technologies/ Configuration Management tools


Methodologies/
Tools

Software Engineering Principles / Chapter 11 / 6 of 26


Lessons Learnt
 Configuration management is a complex activity.
 Implementing configuration management is an iterative process.
 Configuration management should be tested in a controlled, but
real-life environment in a small scale organization.
 Difficult to identify quantitative data to measure process
performance.
 Existence of an operative configuration management system has
shown to make a positive impact on customer relations.
 Existence of an operative change request database has proven to
be valuable in planning product releases.
 The existence of an operative configuration item library has proven
to simplify the process of reconstructing earlier releases.
Software Engineering Principles / Chapter 11 / 7 of 26
Background Information
 The sub-contractors proposed to base the configuration
management procedures on the SPICE Support Process
Category (SUP).

 The project partners agreed and used a process that is called


SUP-2 – Perform Configuration Management .

Software Engineering Principles / Chapter 11 / 8 of 26


Using SUP-2 – The Purpose
 Establish and maintain the integrity of all the products of a
software project throughout the project’s software life cycle

Software Engineering Principles / Chapter 11 / 9 of 26


The SUP-2 Process Tasks
 Establish configuration management library system

 Identify configuration items

 Maintain configuration descriptions

 Manage change requests

 Control changes

 Build product releases

 Maintain configuration item history

 Report configuration status Software Engineering Principles / Chapter 11 / 10 of 26


The ICONMAN Project
Objectives
 Formalize and document internal procedures for configuration
management in each company
 Choose and implement configuration management software
 Produce an internal "Users Guide" for configuration management
describing both the routines and the supporting tools
 Develop metrics to measure the impact of configuration
management
 Develop a network consisting of the participating software
companies and the subcontracted researchers
 Utilize the network for exchange of experiences

Software Engineering Principles / Chapter 11 / 11 of 26


Involved Companies in the
experiment
 Event AS

 Technical Software Consultants (TSC)

 Aktuar Systemer

Software Engineering Principles / Chapter 11 / 12 of 26


Scenario before the
experiment
 Event AS
 All software development is based on one fundamental software

library.
 No formal procedures for handling error reports and requests

for changes from the users


 Technical platform consists of Windows 3.11-workstations in a

Novell Netware 4.x - based LAN.

 TSC AS
 No formal or documented routines for change management of

their main product, FrontSim were in use.


 Procedures for handling error reports were only partially
implemented. Software Engineering Principles / Chapter 11 / 13 of 26
Scenario before the
experiment Contd…
 No formal procedures for handling customer feedback
 Technical environment consists mainly of Unix workstations.

 Aktuar Systemer AS
 The company had implemented a very rudimentary version

control system.
 Simple forms for error reports and change orders were used.

 Co-ordination and assignment of priority to various change

orders was weak.


 No formal test procedures were implemented except for a

limited integration test were used.


 The technical platform consists of Windows95-workstations in a

Novell Netware 4.x - based LAN.

Software Engineering Principles / Chapter 11 / 14 of 26


Work Plan
 First Phase

 Second Phase

 Third Phase

Software Engineering Principles / Chapter 11 / 15 of 26


First Phase
 Lasted 6 months

 Configuration management procedures were developed


 Management of change requests

 Metrics to measure the outcome of the experiment

 Appropriate tools were selected and introduced together with


the procedures.

Software Engineering Principles / Chapter 11 / 16 of 26


Second Phase
 Lasted 7 months

 Consisted of
 Training of the developers

 Refining the configuration management procedures based on

feedback from the developers


 Starting the process of collecting metrics data

Software Engineering Principles / Chapter 11 / 17 of 26


Third Phase
 Lasted 5 months

 Collection of metrics data continued

 Experiment was evaluated

Software Engineering Principles / Chapter 11 / 18 of 26


List of Deliverables
Deliverabl Availibility: Description of the deliverable
e (I)nternal (Title)
Reference (R)estricted
(P)ublic
D1 I Report on definition of CM

D2 I Report on Implementation of
selected tools
D3 R Periodic progress report (first 6
months)
D4 R Report on training on the job

D5 P Mid-term report

D6 R Report on first estimate of the


defined metrics
D7 I Report on refined CM procedures

D8 R Periodic progress reports (6 to 12


months)
D9 P Final Periodic Progress Report

D10 R Consolidated cost statement

D11 P Final Report

Software Engineering Principles / Chapter 11 / 19 of 26


Expected Outcomes
 Improved procedures for handling feedback from customers

 Improved documentation procedures

 Improved testing procedures

 Configuration and change management procedures

Software Engineering Principles / Chapter 11 / 20 of 26


Work Performed
 Organization

 Technical environment

 Training

 Role of the consultants

 Phases of the experiment

 Internal dissemination
Software Engineering Principles / Chapter 11 / 21 of 26
Results and Analysis
 Objectives were met

 The outcome of the experiment corresponded well with the


expected ones

Software Engineering Principles / Chapter 11 / 22 of 26


Objectives after experiment
 Technical

 Business

 Organization

 Culture

 Skills

Software Engineering Principles / Chapter 11 / 23 of 26


Conclusion

 The implementation of configuration management in the three


organizations has been successfully completed.

Software Engineering Principles / Chapter 11 / 24 of 26


Future Actions
 Use of all routines and tools has to be consolidated.

 Further metrics to extend the control over the development


process have to be implemented.

 Documentation of the development process has to be further


improved.

Software Engineering Principles / Chapter 11 / 25 of 26


Summary
 The ICONMAN project’s main objective was to implement
configuration management in three small software development
companies and to assess the effect of this measure from a business
and a technical point of view.
 Configuration management is a complex process and highly
integrated with change management, both with respect to error
correction and handling customer requests.
 Implementing configuration management is an iterative process,
and requires continuous refinement.
 Configuration management in small companies dealing mainly with
the development of one product has a large impact on the way the
whole business is organized.
 Formalized routines for configuration management will help small
companies in planning their production process.
Software Engineering Principles / Chapter 11 / 26 of 26

You might also like