Sustainable and scalable testing strategy in a multilayered service-based architecture
Abstract: This thesis examines and evaluates whether it is possible to measure the quality of a software project and introduces a metric that will evaluate if the quality of every test performed when developing software can be measured. This subject is examined because it can be hard to conclude how well a project and all its parts performed, both during implementation and after it is done. To facilitate this need, this thesis provides a possible solution. To try and answer these questions and meet the needs of this tool a prototype has been developed. The prototype is automated and runs through all the software development tests selected for this project. It sums up the test results and later translates them with the help of a metric to tell its quality grade. The metric is calculated with the help of an arbitrary formula developed for this thesis. Once the metric is concluded the development team working with the project will have an overview of how well each test area is performing and how well the project's end result was. With the help of this metric it is also easier to see if the quality achieved meets the company’s standards and the customer’s wishes. The prototype aims to be sustainable because the solution should last for a long term and also because sustainability means a smoother and more efficient way for developers and other people involved to work with the prototype since not much extra work will be required when updates need to be implemented or other necessary implementations. The prototype is applied on a second project, which is larger and more advanced than the project created for this thesis, to get a better and accurate understanding if the implementation is correct and if the metric can be used as a value to describe a project. The metric results are compared and evaluated. The results of this thesis conclude a proof-of-concept and can be seen as a first step in a longer evaluation and process in determining the quality of tests. The results conclusion is that more parameters and more weighing of each tests importance are needed in order to achieve a reliable metric result. This tool is meant to ease and help developers to quickly come to a conclusion about how good the work is. It could also be beneficial for a company with focus on web development and IT-solutions though it will be easier to follow and set a standard for the services they provide.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)