The transition toward decentralized, data-driven energy systems has intensified the need for transparent and reliable service operations within digital infrastructures such as data spaces. This thesis addresses the challenge of ensuring service-level observability in the EDDIE data space–a modular framework for secure, user-centric energy data sharing. To build trust between users, service providers, and infrastructure operators, a monitoring framework is proposed that evaluates the performance of services through technical Service Level Agreements (SLAs). The framework integrates host-level and service-level monitoring using Prometheus and Spring Micrometer, offering structured metric collection and real-time SLA compliance evaluation. OpenSLO is employed to define Service Level Objectives (SLOs) and alert policies in a machine-readable format, enabling standardized and automated performance verification. Additionally, a two-way SLA mechanism is introduced to ensure that not only third-party services but also AIIDA itself meet performance expectations. The implementation includes a containerized monitoring stack, SLA parsing and evaluation services, and visual dashboards for feedback. The resulting system allows end users and the EDDIE Marketplace to assess service quality dynamically, fostering a transparent and trustworthy data ecosystem. This work contributes both a conceptual model and a functioning prototype for SLA-driven observability in energy-oriented data spaces and offers a foundation for future integration of adaptive, scalable service governance.
Trust Through Metrics: Service Observability and SLA Monitoring in a Data Space
Zauner, M. (Author). 2025
Student thesis: Master's Thesis