Monitoring and Observability in Edge Computing Systems: Taxonomy, Comparative Analysis, and Research Directions

Edge computing is an emerging model for latency-sensitive and distributed applications. However, the observability of edge computing systems in heterogeneous environments remains a challenge, as most existing approaches are limited to only the system, service, application, and network layers. This paper surveys state-of-the-art solutions for edge observability and monitoring. The paper further introduces a thematic taxonomy that groups the state-of-the-art edge observability and monitoring literature based on monitoring intent, telemetry indicators, observability scope, architectural layers, deployment environments, and observability toolchains. Finally, we compare representative solutions in terms of latency, system overhead, bandwidth consumption, and detection accuracy. Our analysis reveals the trade-offs present in existing solutions regarding the overhead and performance metrics, as well as the limitations they impose when scaling, addressing mobility, or augmenting telemetry. Lastly, we conclude with a discussion on open challenges and future directions for designing next-generation, edge observability frameworks.