The paper presents an industrial application of a DevOps process for a Tax fraud detection system. In particular, we report the influence of the quality assessment during development iterations, with special focus on the fulfillment of performance requirements. We investigated how to guarantee quality requirements in a process iteration while new functionalities are added. The experience has been carried out by practitioners and academics in the context of a project for improving quality of data intensive applications.

The paper presents an industrial application of a DevOps process for a Tax fraud detection system. In particular, we report the influence of the quality assessment during development iterations, with special focus on the fulfillment of performance requirements. We investigated how to guarantee quality requirements in a process iteration while new functionalities are added. The experience has been carried out by practitioners and academics in the context of a project for improving quality of data intensive applications.

With the onset of Big Data and Data-Intensive Applications (DIAs) exploiting such big data, the problem of offering privacy guarantees to data owners becomes crucial, even more so with the emergence of DevOps development strategies where speed is paramount. This paper outlines this complex scenario and the challenges therein. On one hand, we outline a tool prototype that addresses the key challenge we found in industry, more specifically, assisting the process of continuous DIA architecting for the purpose of offering privacy-by-design guarantees. On the other hand we define a research roadmap in pursuit of a more correct and complete solution for ensured privacy-by-design in the context of Big Data DevOps.