| Because of the extensive application of computer, the research on software reliability is very important. In order to get the reliability of software, we should do software reliability evaluation. The software reliability evaluation, based on software reliability models, using data collected from software testing, can provide the numerical measurement of reliability.In order to evaluate the reliability of software, software testing is necessary to collect data for software reliability models. The dissertation describes the steps of testing and the concrete methods based on Monitor System, and summarizes some useful experiences.People doing research on the reliability of the software through decades , has raised more than 100 kinds of software reliability model. These models use the fault data produced at the testing phase that can give indicators of the software reliability measure with quantitative methods. Thus it can assistant software developers and managers to make the appropriate decision-making according to software reliability evaluation , which has great significance in improving the quality of the software .This paper is an study of Software Testing Technology based on practicing; It selects software reliability evaluation model, and gets indicators of the reliability measure of model-corresponding; And it achieve the algorithm which comes down to the model of software reliability evaluation ;And it achieve "the system of software reliability evaluation and testing-assisted ."There are eight software reliability models in the implemented Software Reliability Evaluation and Auxiliary Testing System: Two-Argument G-ONHP Model,Three-Argument G-ONHPP Model, J-M Model, Weibull Model, GM Model YOM Model, NHP3AD Model and DS Model. The dissertation provides calculating methods of reliability measurement for each model.Finally, The fifth chapter of this paper describes the implementation of the Software Reliability Evaluation and Auxiliary Testing System. |
PDF Full Text Request