A scalable software quality verifier framework

Abstract

Outsourcing software development is a growing business that is proven to bring cost-effective and efficient solutions for varying demands of software product companies. Though it has proven its capability in bringing value to products to stay ahead in competition, few inherent problems are also identified in this practice. A prominent issue is how to verify the quality of the applications delivered by the vendor. Given that a critical bug in production can bring disasters, it is vital to the outsourcer to make sure that the deliverables from the vendor conform to a well defined set of quality guidelines. The work described here is the design and implementation of a scalable software quality verification framework on top of which, industrial grade automated quality verification systems can be built with minimum effort.

The framework is built to evaluate both software code and applications. Code level evaluation is done in two phases; when the developer tries to add code to the repository and a deeper test covering a wide range of problems in an offline context. The rules used for evaluation, actions on results and alerting can be customized in project level.

The framework provides a programming interface and a set of tools for application evaluation. The simple yet powerful programming interface creates ground for building a knowledgebase accumulating the experience of veterans. This is used in collaboration with modern tools to evaluate applications against their performance, security, memory and 10 usage, etc.

A quality verification system built using the framework which was put into action in a commercial software project proved to add a significant value to the deliverables. An experiment done with the programming interface showed that powerful analysis systems can be built to both evaluate deliverables and aid in software due-diligence process.