Abstract: Although, reliability is an important attribute of quality, especially for mission critical systems, yet, there does not exist any versatile model even today for the reliability assessment of component-based software. The existing Black Box models are found to make various assumptions which may not always be realistic and may be quite contrary to the actual behaviour of software. They focus on observing the manner in which the system behaves without considering the structure of the system, the components composing the system, their interconnections, dependencies, usage frequencies, etc.As a result, the entropy (uncertainty) in assessment using these models is much high.Though, there are some models based on operation profile yet sometimes it becomes extremely difficult to obtain the exact operation profile concerned with a given operation. This paper discusses the drawbacks, deficiencies and limitations of Black Box approaches from the perspective of various authors and finally proposes a conceptual model for the reliability assessment of software.
Abstract: Accurate software reliability prediction not only enables developers to improve the quality of software but also provides useful information to help them for planning valuable resources. This paper examines the performance of three well-known data mining techniques (CART, TreeNet and Random Forest) for predicting software reliability. We evaluate and compare the performance of proposed models with Cascade Correlation Neural Network (CCNN) using sixteen empirical databases from the Data and Analysis Center for Software. The goal of our study is to help project managers to concentrate their testing efforts to minimize the software failures in order to improve the reliability of the software systems. Two performance measures, Normalized Root Mean Squared Error (NRMSE) and Mean Absolute Errors (MAE), illustrate that CART model is accurate than the models predicted using Random Forest, TreeNet and CCNN in all datasets used in our study. Finally, we conclude that such methods can help in reliability prediction using real-life failure datasets.
Abstract: Measuring the reusability of Object-Oriented (OO) program code is important to ensure a successful and timely adaptation and integration of the reused code in new software projects. It has become even more relevant with the availability of huge amounts of open-source projects. Reuse saves cost, increases the speed of development and improves software reliability. Measuring this reusability is not s straight forward process due to the variety of metrics and qualities linked to software reuse and the lack of comprehensive empirical studies to support the proposed metrics or models. In this paper, a conceptual model is proposed to measure the reusability of OO program code. A comprehensive set of metrics is used to compute the most significant factors of reusability and an empirical investigation is conducted to measure the reusability of the classes of randomly selected open-source Java projects. Additionally, the impact of using inner and anonymous classes on the reusability of their enclosing classes is assessed. The results obtained are thoroughly analyzed to identify the factors behind lack of reusability in open-source OO program code and the impact of nesting on it.
Abstract: Software reliability prediction gives a great opportunity to measure the software failure rate at any point throughout system test. A software reliability prediction model provides with the technique for improving reliability. Software reliability is very important factor for estimating overall system reliability, which depends on the individual component reliabilities. It differs from hardware reliability in that it reflects the design perfection. Main reason of software reliability problems is high complexity of software. Various approaches can be used to improve the reliability of software. We focus on software reliability model in this article, assuming that there is a time redundancy, the value of which (the number of repeated transmission of basic blocks) can be an optimization parameter. We consider given mathematical model in the assumption that in the system may occur not only irreversible failures, but also a failure that can be taken as self-repairing failures that significantly affect the reliability and accuracy of information transfer. Main task of the given paper is to find a time distribution function (DF) of instructions sequence transmission, which consists of random number of basic blocks. We consider the system software unreliable; the time between adjacent failures has exponential distribution.
Abstract: This paper presents a computational methodology
based on matrix operations for a computer based solution to the
problem of performance analysis of software reliability models
(SRMs). A set of seven comparison criteria have been formulated to
rank various non-homogenous Poisson process software reliability
models proposed during the past 30 years to estimate software
reliability measures such as the number of remaining faults, software
failure rate, and software reliability. Selection of optimal SRM for
use in a particular case has been an area of interest for researchers in
the field of software reliability. Tools and techniques for software
reliability model selection found in the literature cannot be used with
high level of confidence as they use a limited number of model
selection criteria. A real data set of middle size software project from
published papers has been used for demonstration of matrix method.
The result of this study will be a ranking of SRMs based on the
Permanent value of the criteria matrix formed for each model based
on the comparison criteria. The software reliability model with
highest value of the Permanent is ranked at number – 1 and so on.
Abstract: Software Reliability is one of the key factors in the software development process. Software Reliability is estimated using reliability models based on Non Homogenous Poisson Process. In most of the literature the Software Reliability is predicted only in testing phase. So it leads to wrong decision-making concept. In this paper, two Software Reliability concepts, testing and operational phase are studied in detail. Using S-Shaped Software Reliability Growth Model (SRGM) and Exponential SRGM, the testing and operational reliability values are obtained. Finally two reliability values are compared and optimal release time is investigated.
Abstract: Testing accounts for the major percentage of technical
contribution in the software development process. Typically, it
consumes more than 50 percent of the total cost of developing a
piece of software. The selection of software tests is a very important
activity within this process to ensure the software reliability
requirements are met. Generally tests are run to achieve maximum
coverage of the software code and very little attention is given to the
achieved reliability of the software. Using an existing methodology,
this paper describes how to use Bayesian Belief Networks (BBNs) to
select unit tests based on their contribution to the reliability of the
module under consideration. In particular the work examines how the
approach can enhance test-first development by assessing the quality
of test suites resulting from this development methodology and
providing insight into additional tests that can significantly reduce
the achieved reliability. In this way the method can produce an
optimal selection of inputs and the order in which the tests are
executed to maximize the software reliability. To illustrate this
approach, a belief network is constructed for a modern software
system incorporating the expert opinion, expressed through
probabilities of the relative quality of the elements of the software,
and the potential effectiveness of the software tests. The steps
involved in constructing the Bayesian Network are explained as is a
method to allow for the test suite resulting from test-driven
development.