An MQTT-based Resource Management Framework for Edge Computing Systems

Abstract

The complexity of IoT systems and tasks that are put before them require shifts in the way resources and service provisioning are managed. The concept of edge computing is introduced to enhance IoT systems' scalability, reactivity, efficiency, and privacy. In this paper, we present an edge computing solution for resource management of context-aware decision-making processes distributed between IoT gateways. The solution performs decision-making process management for smart actuation, based on analysis of sensory data streams, and context-informed edge computing resource and service provisioning management based on topology and operational changes. Our architectural solution showcases the first version of a Resource Management Framework - a generic framework for software resource orchestration best-suited to IoT platforms with event-driven, publish-subscribe communication mechanisms. Proof of concept experiments that are executed in a simulated edge computing testbed validate our solution's performance in improving the resilience and responsiveness of the edge computing system when there are operational and topology changes. Furthermore, the framework addresses the recovery of failed decision-making processes, impacting the overall health of the underlying IoT system.