Testing Approaches For Component Based Software Systems: A Systematic Review (Research Paper Sample)

TESTING APPROACHES FOR COMPONENT BASED SOFTWARE SYSTEMS: A SYSTEMATIC REVIEWResearch Survey Student name[Date]Roll no.
“Testing Approaches for Component Based Software Systems: A Systematic Review”
Abstract— Component-based software engineering (CBSE) has emerged as a methodology that offers quick improvement of framework utilizing less assets and effort. If the components offer dependable services, the core idea of reuse and cutting down the improvement expenses can be accomplished. In this way, incorporation and testing of components become an important phase in CBSE. Incorporation or integration of components is a very significant activity in CBSE. This includes understanding correspondence and coordination between the components. Developers are not provided with sufficient information on these components. As a consequence, understanding data flow while incorporating these components is a challenge. Component-based software facilitates development of complex systems by allowing integration of reusable components. Testing components is a challenging area of exploration. There have been troubles integrating the components. This thus influences the quality and reliability of the software. Our research aims at finding the existing testing techniques and processes in Component Based Software System (CBS). The systematic literature survey is based on 7 articles gathered from numerous stage determination process. These articles have been published within the time span of 2010-2016.
Keywords—Testing Approaches, Component Based Testing, Component Testing Process, Testing Strategies, Systematic Review.
INTRODUCTION
Component Based Software System (CBS) are primarily built from reusable components such as third party components and Commercial-Of-The-Shelf (COTS) components. Because of this, component based systems are developed rapidly with least engineering efforts and resource cost. Component Based Software Engineering covers many software engineering disciplines and different practices. Many of them have been developed and successfully implemented. CBSE has been successful in certain designing areas, for example, distributed web based applications and office applications however it is still in the early phase of use in numerous different spaces, specifically those which have particular prerequisites on various quality properties.
Component-Based Development (CBD) offers a drastically new way to deal with the design, development, execution and advancement of programming applications. Programming applications are gathered from components from a variety of sources; the components might be composed in a few diverse programming languages and keep running on a few distinct platforms or we can say that components are heterogeneous in nature.
To ensure the delivery of quality component based software, efficient and proficient testing is the key process in software development. Component Based Software has implementation transparency property and its incessant heterogeneity raise a few troubles when the product is to be tried. Since segments are proposed to be reused crosswise over different items and item families, components must be tried sufficiently.
Software component testing refers to testing that analyzes component alongside its outline, produces part tests, distinguishes segment blames and assesses component unwavering quality. So component testing plays very important role for the development of a quality component based software product. There are numerous testing procedures that are useful in component based testing in light of the fact that these are valuable to create data about the components and assess the validity of the part for the new software products or applications.
Graphical user interfaces (GUIs) are by far the most popular means used to interact with today’s software The majority of the product's being utilized today is GUI based and the dependability and power of the product relies on upon how every part works in as GUI is a standout amongst the most critical segment so it should be tried for practical rightness.
The attributes of the GUI are unique in relation to that of routine programming as it is totally occasion based and progressive in nature. So an alternate representation "event forest" is made which is totally fit for re-enacting the various levelled nature of the GUI furthermore stores the occasion cooperation data.
Most of the software’s today have GUI front-end and the newer versions are released year by year. The newer versions have modifications as compared to previous version. So to test the modified GUI, regression testing is performed which reuses the test cases from the old GUI.
The rest of the paper is organized as follows. In section 2, we describe the method we followed for systematic survey. It includes our research questions, search strategy, inclusion/exclusion criteria in selection of the articles, and quality assessment of the articles. In Section 3 we discuss our results. Conclusions are presented in Section 4.
RESEARCH METHOD
In doing the systematic literature survey we followed the guidelines provided by researchers in ADDIN EN.CITE J. Biolchini20056[1]6617J. Biolchini, P.G. Mian, A.C. Natali, G.H. TravassosSystematic Review in Software Engineering: Relevance and UtilityTechnical Report ES-679/05, PESC-COPPE/UFRJ Available at Report ES-679/05, PESC-COPPE/UFRJ Available at [1]. According to ADDIN EN.CITE J. Biolchini20056[1]6617J. Biolchini, P.G. Mian, A.C. Natali, G.H. TravassosSystematic Review in Software Engineering: Relevance and UtilityTechnical Report ES-679/05, PESC-COPPE/UFRJ Available at Report ES-679/05, PESC-COPPE/UFRJ Available at [1], a systematic literature survey is a systematic process of finding and assessing the available research related to particular questions of interest. An important reason for undertaking such a study is to identify areas of further investigation/research.
A. Research Questions we formulated the following objectives.
1 Survey of existing testing techniques for CBS.
2 Understand Component Testing Process.
3 Compare the existing techniques and find scope for overcoming the challenges.
4 Identify the issues in determining the testability of components.
5 A document containing a formal report of the above mentioned objectives.
To get good understanding of the above mentioned, we formulated research questions. These are as follows.
1 What is the state of art in testing of software components?
2 What are the different phases in testing CBS?
3 What are the existing techniques?
4 How does component testing process work?
5 What is the effectiveness of component testing process?
6 How is testability of components determined?
B. Search Strategy
Our search strategy is based on the following.
1) Identification of search terms to find articles that relate to our research questions. The search terms alternate words and synonyms. The following search terms are used in our search strategy – component-based software, component
based software system testing techniques, COTS, components, challenges, testing challenges, testability, components behaviour(u)r, testing process, component testing process.
2) Use of Boolean AND to join the search terms and use of Boolean OR to join synonyms/alternate words.
3) We used a multiple-phase strategy for searching. Firstly, we searched electronic databases. We chose the 2010 as the starting year of search. We searched within the following electronic database:
- ACM Digital Library.
To limit the number of articles in every search, we selected the articles available in full text. We then selected the articles based on abstract. In the second stage, we scanned references and selected a few that were available in the above mentioned journal only.
C. Inclusion/Exclusion criteria
CBS is a vast field of research. To constrain our exploration to software testing, i.e. as per our exploration questions we defined criteria to incorporate and avoid articles in our study. We connected the accompanying consideration criteria.
- Published within the time span of 2010 – 2016
- Article available in full text
- Articles can be on controlled experiments, literature survey or industrial practice
- Articles on models or frameworks for testing
- Articles evaluating on existing models or extending the models
We applied the following exclusion criteria.
- Articles not related to Software Engineering/Computer Science
- Articles that do not match the inclusion criteria
- Articles not related to Software Testing
- Articles on test plan
- Articles on Metrics
- Articles on Maintenance
- Articles on Service oriented software systems, Object Oriented software systems
- Articles on software testing tools
D. Quality assessment
Quality data is essential in synthesis of the results ADDIN EN.CITE J. Biolchini20056[1]6617J. Biolchini, P.G. Mian, A.C. Natali, G.H. TravassosSystematic Review in Software Engineering: Relevance and UtilityTechnical Report ES-679/05, PESC-COPPE/UFRJ Available at Report ES-679/05, PESC-COPPE/UFRJ Available at [1]. We applied the following criteria to select studies of good quality.
- Is the reader able to understand the aims of the research?
- Is the context of study clearly stated?
- Is the study verified/ assumptions proved?
-Is the research useful to software industry/research community?
- Are the conclusions related to the aim and the very purpose of the goal of the research?
RESULTS
In this section, we describe the results of our search. We classified the results into three parts. These are:
* Testing techniques
* Testing process
* Testability
Tables 1, 2 and 3 show the distribution of papers pertaining to the above classi...