This deliverable is the result of Task 3.1: Trustworthiness Framework and Metrics, whose goal was to design the overall trustworthiness monitoring and assessment framework. The task
started with a detailed analysis of the trustworthiness properties of interest for different layers of the ATMOSPHERE ecosystem, from which the set of trust scores will be designed. Taking into account the selected properties and metrics, the trustworthiness monitoring and assessment lifecycle was defined, detailing the sequence of stages that trust information should undergo, from monitoring to measurement, ending with analysis for decision making (e.g. runtime adaptation). To support the defined lifecycle, the requirements of the monitoring platform and of the trustworthiness measurement and analysis services were specified.

In practice, this deliverable defines the requirements for the assessment of trustworthiness of cloud services dealing with data and supporting the development of more trustworthy cloud applications, which are the concerns of the ATMOSPHERE project. Four aspects are addressed:

  • The trustworthiness assessment lifecycle, which defines the sequence of stages that should be considered for trustworthiness assessment, including both design‑time and runtime assessment and improvement/adaptation;
  • A preliminary definition of the trustworthiness properties of interest regarding the different layers of the ATMOSPHERE ecosystem, including the identification of potential scores and a discussion on the use of quality models for the implementation of the trustworthiness scores;
  • The requirements of the trustworthiness assessment platform, focusing on both the functional and non‑functional requirements that should be considered for the implementation of the monitoring and assessment solution;
  • An high‑level architecture of that same platform.

The deliverable moves forward the state‑of‑the‑art on trustworthiness characterization in the sense that it proposes a generic trustworthiness lifecycle, defines and discusses different
trustworthiness properties applicable to multiple scenarios, and shows how quality models can be built and used in practice to compute scores. These three contributions are of key relevance in the context of ATMOSPHERE, but can also be applied in other contexts. In practice, although focused on the project needs, the deliverable is of use for researchers and practioners working on the design/development of trustworthy cloud‑based systems, in general.