TOWARD A QUALITY MODEL FOR CBSE

Conceptual Model Proposal

María A. Reyes, Maryoly Ortega, María Pérez, Anna Grimán

Luis E. Mendoza and Kenyer Domínguez

Departamento de Procesos y Sistemas, Universidad Simón Bolívar, P.O. Box 89000, Caracas 1080-A, Venezuela

Keywords: Quality model, Component-based software engineering, Components.

Abstract: In this paper, which is part of a research in progress, we analyze the conceptual elements behind

Component-Based Software Engineering (CBSE) and propose a model that will support its quality

evaluation. The conceptual model proposed integrates the product perspective, a view that includes

components and Component-Based Software (CBS), as well as the process perspective, a view that

represents the component and CBS development life cycle. The model proposal was developed under a

systemic approach that will allow for assessing and improving products and processes immersed in CBSE.

Future actions include proposing metrics to operationalize the model and validate them through a case

study. The model application will allow studying the behavior of each perspective and the relationships

among them.

1 INTRODUCTION

As software engineering has developed, new

techniques and tools have been employed to improve

development processes. Likewise, engineering scope

has been expanded and sub-disciplines have been

produced that serve as a framework for the software

development process. In this regard, Component-

Based Software Engineering (CBSE) is focused on

the design and construction of applications that use

software components; it is oriented toward the

acquisition of existing components, and takes into

account the interdependence between components

and these applications (Pressman, 2005; Clemente

and Hernández, 2003). In this regard, CBSE shares

common grounds with Service Oriented Architecture

(SOA) in its interest in developing organizational

relationships: develop partnerships between

developers and users; develop key processes. Mutual

support between the application generator and

service provider communities is necessary

(Anderson, 2007).

One of the CBSE processes is the development

of components as reusable entities for systems that

requires applying established methodologies.

Another process is the CBS development, which

includes maintenance and improvement of systems

through personalization and replacement of

components (Councill and Heineman, 2001;

Crnkovic, 2003).

CBSE objectives allow us to identify two

perspectives within CBSE itself: the product

perspective, a view that includes software

components and CBS, and the process perspective, a

view that represents the component and CBS

development life cycle. However, for applications

developed in CBSE to succeed, it is necessary to

consider quality aspects, oriented to the identified

perspectives. In this regard, it is possible to obtain a

model that includes characteristics and sub-

characteristics for each perspective through a quality

model that provides a set of guidelines aimed at

specifying quality requirements and assessment.

In this paper we analyze the conceptual elements

behind Component-Based Software Engineering

(CBSE) and propose a model that will support its

quality evaluation. By using a systemic approach,

our proposal integrates the product perspective,

which includes software components and CBS, and

the process perspective that includes how software

components and CBS are constructed.

The remainder of this paper is structured as

follows. In section 2 we discuss the background for

this research. Then, in section 3, we give a brief

description of the methodology followed in this

work. Afterwards, in section 4 we describe each sub-

101

Reyes M., Ortega M., Pérez M., Grimán A., Mendoza L. and Domínguez K. (2009).

TOWARD A QUALITY MODEL FOR CBSE - Conceptual Model Proposal.

In Proceedings of the 11th International Conference on Enterprise Information Systems - Databases and Information Systems Integration, pages

101-106

DOI: 10.5220/0001978701010106

 SciTePress

model and the integrated model. Finally, section 5

provides the conclusions and future works.

2 BACKGROUND

CBSE emerges as a reuse approach for software

Systems and it is oriented toward the design and

construction of systems using software components

(Pressman, 2005). The ability to construct full

solutions interconnected through interfaces created

with components is one important factor to consider

within CBSE.

For the purposes of this research, we define

component as a software element subject to

composition by third parties, which can be used by

other software elements and conforms with a

component model (SEI, 2000; Meyer, 2003).

Councill and Heineman (2001) point out that

components exhibit a number of characteristics:

those referred to the properties inherent in the

component, which are basic characteristics that

every component must have, and the desirable or

advanced properties of the components, which are

related to the different functions offered by the

component. The desirable characteristics of a

component may be present or not, but this not

jeopardizes its functionality; however, if

components have these characteristics, they can

guarantee the quality of the component as a product.

CBS, in turn, comprises a number of self-

contained component units. Components may have

been developed in different languages, executed on

different platforms and distributed throughout

different geographical locations (Gao et al., 2003).

CBS may be built with its own software

components, which can be specifically developed for

the software or with commercial off-the-shelf

(COTS) acquired for this particular purpose.

Additional to the concepts related to CBSE and

as a function to measure and assess component

quality, some authors have proposed quality models

to assess and select software components (Bertoa et

al., 2006; Simão and Belchior, 2003; Rawashdeh

and Matalkah, 2006; Carvallo, Franch and Quer,

2006; Andreou and Tziakouris, 2007; Carvallo et al.,

2007).

Furthermore, quality models for CBS has been

presented by several authors: Jasmine and Vasantha

(2007) propose a model to measure quality of CBS

products; the Sedigh-Ali and Paul (2001) model is

focused on measuring quality of COTS-based

systems; and finally, Grunske (2007) presents a

generic framework for early prediction of CBS

quality.

All these proposals stress quality assurance

oriented to measuring the properties of the CBS

product; they do not focus on quality of those

aspects related to the life cycle of CBS development

as a whole or consider quality assurance of software

components at the same time.

In this regard, Grunske (2007) highlights the

importance of proposing unified quality assessment

models for CBSE, which include the component

development life cycle and its inclusion into systems

so that to have the proper management of systems on

a large scale, and support technology development

and transference to industrial applications.

3 METHODOLOGY

Since the aim of this research was establishing a

series of concepts behind CBSE (a CBSE conceptual

model), we considered that a methodology for

ontology creation was suitable for our purpose. The

main goal is to present conceptual elements that

allow defining a quality model for CBSE. In fact,

Noy and McGuinness (2001) define ontology as an

explicit and formal description of concepts in a

discourse domain. They propose the following seven

steps for creating an ontology, which were useful in

our research:

1. Determine the domain and scope of the

ontology

2. Consider reusing existing ontologies

3. Enumerate important terms in the ontology

4. Define the classes and the class hierarchy

5. Define the properties of classes

6. Define class relations

7. Create instances

The scope of this paper does not include

detailing each one of the proposed steps; however,

the activities carried out in every step are mentioned:

(1) elements common to each definition, were

identified to find the relevant terms; (2) ontologies

using a product-process approach were not found,

therefore existing ontologies were not reused; (3)

significant terms are presented in the sub-models,

specifying their hierarchical order as well as the

relations among them; (4) properties are not

specified or instances are not created at this level,

because this has still to be applied to a case study.

The outputs of the reminding steps are described in

the following sections.

ICEIS 2009 - International Conference on Enterprise Information Systems

102

4 SUB-MODELS INTEGRATION

The quality model for the CBSE proposed herein

consists of five sub-models established for the

product and process perspectives. The component

and the CBS sub-models are defined within the

product perspective. The sub-models for the CBS

life cycle and for the component life cycle are

defined for the process perspective. It is important to

highlight that in the process perspective, the sub-

models represent process maturity models, which

will allow to measure quality in each process phase.

Finally, we show the Component Model, which

presents standards that allow us to precise a

component, its documentation and implementation,

and also is the backbone of a CBS (Gao et al., 2003),

providing support for development, communication,

evolution, composition and implementation The

Component Model is related with the component

and CBS sub-model.

4.1 Product Perspective Sub-model:

Components

Figure 1: Component sub-model.

The relations established and the elements identified

for the component sub-model consider the

classifications proposed by Councill and Heineman

(2001), Montilva et al., (2003), and Gao et al.,

(2003), who recognize the existence of mandatory

characteristics related to identification,

independence, composition, documentation and

accessibility to the component through its interface.

Concerning the desirable characteristics of a

component, these authors refer to maintainability,

personalization, portability, reliability, and

certification. We classify characteristics as

prioritary, those that must be met by every

component effectively to ensure that the unit

evaluated corresponds to a component, and

complementary, those that add value to the

component and allow us to certify the component

quality. Figure 1 we show the characteristics defined

in the component sub-model that will allow us to

establish metrics to measure component quality.

4.2 Product Perspective Sub-model:

CBS

The quality sub-model to measure those aspects

related to CBS are based on the properties identified

by Gao et al., (2003). These characteristics are

unique and differ from traditional software systems,

highlight the relations existing between the

developed software and the components of which it

consists; they also call for the need of component

interoperability, and promote software and

component reuse and the evolution (maintainability)

of the developed systems, among others.

In Figure 2 we present the CBS quality model

based in ISO/IEC 9126. It can be observed that the

elements proposed correspond to the characteristics

the CBS must comply with, which are related to

components and their integration into the software.

Figure 2: CBS sub-model.

4.3 Process Perspective Sub-model:

CBS Life Cycle

Concerning the life cycle of CBS development, Cai

et al. (2002), Crnkovic (2003), Sommerville (2005),

and Pressman (2005) agree that CBS life cycle

includes stages that are comparable to those in other

processes; however, they consider that it is

fundamental to include elements related to

components and their composition into CBS. Figure

3 shows that CBS development life cycle includes

seven stages that relate the software development

TOWARD A QUALITY MODEL FOR CBSE - Conceptual Model Proposal

103

4.5 Component Model

life cycle to the elements associated to components

and their composition.

According to Sommerville (2005), a components

model consists of a definition of the standards that

allow for precising a component, its documentation

and implementation. Besides, Gao et al. (2003) point

out that the component model is the backbone of a

CBS (e.g.: CORBA, DCOM, JavaBeans). Figure 5

presents the component model, which highlights the

elements making up the components model, as well

as the support for development, communication,

evolution, composition and implementation of

components. This components model is related with

the CBS lifecycle in the composition phase, and with

the component sub-model when the component

complies the components model.

Figure 3: CBS life cycle sub-model.

4.4 Process Perspective Sub-model:

Component Life Cycle

The component life cycle reflects how a component

is constructed, its advantages and disadvantages,

how it is integrated into a system, quality aspects

and metrics required to assess process quality itself.

Cai et al., (2002) and Lau and Wang (2006) identify

three stages or phases in the life cycle: design,

implementation and execution environment of the

component. Figure 4 shows the phases identified for

the component lifecycle: analysis of component

requirements, development of the component within

a system, and certification and personalization,

viewed through the implementation and

modification of the component for it to be adjusted

to the requirements inherent in the system they are

going to be integrated.

Figure 5: Component Model.

Once we followed the methodology for defining

sub-models, we applied a method proposed by

(Carvallo et al., 2004) to define and integrate quality

models for CBSE. As a result, we obtained the

conceptual model shown in Figure 6, which is based

on integrated concepts and presents the relations not

shown in the sub-models identified in CBSE. The

relations between components and the component

lifecycle, between CBS and its lifecycle; finally the

relations between component models, components

and CBS development are presented.

This global view of concepts also allows us to

identify the important role of the sub-model named

components model which depicts properties to be

considered during the whole life cycle of the CBS. It

is to be one of the main elements in any CBS quality

evaluation model. This way, both perspectives are

closely related to each other and that they are

mutually affected. Notice that quality of a

component directly influences CBS quality, through

Figure 4: Component lifecycle sub-model.

ICEIS 2009 - International Conference on Enterprise Information Systems

104

Figure 6: Integrated conceptual model for CBSE.

components models that are the backbone of these

software systems. Furthermore, software

construction, in each one of its phases, contemplates

those component-related aspects that range from the

component selection through their composition into

systems and their maintenance or replacement.

5 CONCLUSIONS AND

FUTURE WORK

We analyzed and presented a model that integrates

the conceptual elements behind the formulation of a

quality model for CBSE. From a systemic point of

view, the model will allow us to assess and certify

the product and process perspectives identified. Our

proposal establishes and defines the relationships

between products (components and CBS), and

processes, related to how components and CBS are

constructed. Furthermore, we can highlight two

findings: 1) a close relationship exist between both

identified perspectives: quality of a component

directly influences CBS quality, 2) the component

models are the backbone of these software systems.

Future works within the scope of this research-

in-progress includes developing the metrics required

to effectively measure all the aspects proposed in the

integrated quality model. A GNU/Linux based

distribution will be used for the case study on which

the quality model is to be applied.

ACKNOWLEDGEMENTS

This research was partially supported by the

National Fund of Science, Technology and

Innovation, Venezuela (FONACIT), number G-

2005000165 and project USB/DID S1-IN-CAI-

01206.

REFERENCES

Anderson, W. 2007. What COTS and Software Reuse

Teach us about SOA. In Sixth International IEEE

Conference on Commercial-off-the-Shelf (COTS)-

Based Software Systems. 141-149.

TOWARD A QUALITY MODEL FOR CBSE - Conceptual Model Proposal

105

Andreou, A., Tziakouris, M., 2007. A quality framework

for developing and evaluating original software

components. In Information and Software Technology,

49, 122–141.

Bertoa, M. F., Troya, J. M. and Vallecillo, A., 2006.

Measuring the usability of software components. In

Journal of Systems and Software, 79(3), 427-439.

Cai, X., Lyu, M., Wong, K., 2002. Component-Based

Embedded Software Engineering: Development

Framework, Quality Assurance and a Generic

assessment environment. In International Journal of

Software Engineering and Knowledge Engineering,

12(2), 107-133.

Carvallo, J., Franch, X., Quer, C., 2006. Managing Non-

Technical Requirements in COTS Components

Selection. In 14

IEEE International Requirements

Engineering Conference, 316 - 321.

Carvallo, J., Franch, X., Quer, C., 2007. Determining

Criteria for Selecting Software Components: Lessons

Learned. In IEEE Software, 24(3), 84-94.

Carvallo, J., Franch, X., Grau, G., Quer, C., 2004.

COSTUME: A Method for Building Quality Models

for Composite COTS-based Software Systems. In

Fourth International Conference on Quality Software,

Germany. IEEE Computer Society Press, 214-223.

Clemente, P., Hernández, J., 2003. Aspect Component

Based Software Engineering. In Second AOSD

Workshop on Aspects, Components, and Patterns for

Infrastructure Software.

Councill, W., Heineman, G., 2001. Component-Based

Software Engineering. Addison Wesley.

Crnkovic, I., 2003. Component-based Software

Engineering – New Challenges in Software

Development. In 25th International Conference on

Information Technology Interfaces.

Gao, J., Tsao, J., Wu, Y., 2003. Testing and Quality

Assurance for Component-Based Software. Artech

House Publishers.

Grunske, L., 2007. Early quality prediction of component-

based systems – A generic framework. In The Journal

of Systems and Software, 80, 678–686.

Jasmine, K y Vasantha, R., 2007. DRE - A Quality Metric

for Component based Software Products. In

Proceedings of World Academy of Science,

Engineering and Technology, 23, 380-383. Lau, K.,

Wang, Z., 2006. A Survey of Software Component

Models. University of Manchester: Computer Science.

Meyer, B., 2003. The Grand Challenge of Trusted

Components. In 25

International Conference on

Software Engineering.

Montilva, J., Arapé, N., Colmenares, J., 2003. Desarrollo

de Software Basado en Componentes. In IV Congreso

de Automatización y Control, Mérida.

Noy, N. y McGuinness, D., 2001. Ontology Development

101: A Guide to Creating Your First Ontology.

Stanford University: Stanford, CA.

Pressman, R., 2005. Software Engineering: A

Practitioner’s Approach,. McGraw Hill. 6

. edition.

Rawashdeh, A. y Matalkah, B., 2006. A New Software

Quality Model for Evaluating COTS Components. In

Journal of Computer Science 2(4), 373-381.

Sedigh-Ali, S., y Paul, R., 2001. A Software Engineering

Metrics for COTS-Based Systems. In Computer, 44-

50.

SEI, Software Engineering Institute, 2000. Volume II:

Technical Concepts of Component-Based Software

Engineering. Carnegie Mellon University. 2

edition.

Simão, R, Belchior, A. 2003. Quality characteristics for

software components: Hierarchy and quality guides. In

Lecture Notes in Computer Science, 2693, 184-206.

Sommerville, I., 2005. Software Engineering. Addison

Wesley. 7

edition.

Szyperski, C., 2002. Component Software Beyond Object-

Oriented Programming. Addison Wesley. 2

edition.

ICEIS 2009 - International Conference on Enterprise Information Systems

106