Software Dependability Assessment Using DevOps Metrics

Assessing software dependability is a major task of Total Management Dependability strongly motivated by the framework Total Quality Management and Lean Six Sigma methodology used by software organizations to improve performance and assuring constant collaboration with stakeholders. The most recent approach in improving software quality is DevOps, a merging of software development, software quality assurance, and software deployment and integration (IT operations). This paper addresses DevOps metrics useful to assess the software dependability. Both static and dynamic models are considered. Firstly, recent developments on software reliability, availability, safety, security, and resilience are reviewed. In the second part, extensions of dependability metrics obtained by soft computing (fuzzy, intuitionistic-fuzzy, and neutrosophic numbers) are presented along with practical examples.