A Continuous Certification Methodology for DevOps

Abstract

The cloud paradigm has revolutionized the way in which software systems are designed, managed, and maintained. With the advent of the microservice architecture, this trend was brought to the extreme, pushing the whole software development process towards unification of software development (Dev) and software operation (Ops). This rapid evolution has not immediately found counterparts in assurance techniques, where the evaluation of the non-functional behavior of a software system and of the software development process are completely decoupled. In this paper, we put forward the idea that next-generation assurance techniques, and more specifically certification techniques, must evaluate a software system throughout the whole development process. To this aim, we define a continuous certification scheme for DevOps that evaluates the software artifacts produced at each stage of the development process. We then present the assurance framework managing our certification scheme and experimentally evaluate the continuous certification scheme in a real DevOps scenario. CCS CONCEPTS• Software and its engineering → Software verification and validation; Extra-functional properties; Development frameworks and environments; • Security and privacy → Software security engineering.

Microservice security: a systematic literature review

“…However, the majority of the selected publications provide no in-depth security analysis of any of the two development approaches, but rather indicate the inclusion of generic security measures in the steps of the development method. Only three works, namely Mansfield-Devine (2018) , Anisetti et al (2019) and Kumar & Goyal (2020) , propose concrete and specific variants of the DevOps methodology that tackle security issues—in particular Mansfield-Devine (2018) explicitly cites the guidelines of DevSecOps Hsu (2018) .…”

Security Assurance in Modern IoT Systems

“…For instance, our experiments uncovered several relevant issues without stressing the system execution. In this context, DevSecOps stands out as a cornerstone, to the point that assurance evaluations can become a part of a DevSecOps pipeline [22]. At the same time, the system must be evaluated at all layers, possibly exploiting the edge layer to (indirectly) assess IoT nodes without impacting on their resources.…”

“…These development artifacts describe the complete system deployment, and therefore become relevant for assurance, in terms of requirements to verify at run time, and of targets of evaluation. For instance, an assurance evaluation for availability can be executed against the manifest looking for High-Availability policies, in addition or in replacement of being executed against the system [22]. This brings crucial advantages when the system cannot be assessed directly (e.g., because of resource constraints), and constitutes a complementary and less-invasive means of evaluation.…”

Bridging the Gap Between Certification and Software Development

“…Non-functional properties mimics Protection Profiles of CC [25]. They have been mostly considered in the context of service certification [2] and later enhanced in the context of cloud certification [3,34,43], cloud plan adaption [6], DevOps pipelines [4], and to complement and validate risk management [1].…”