Evaluating Distributed MQTT Brokers' Performance for Scalable Augmented Reality IoT Data Visualization based on Collective Intelligence

Recent technological and scientific trends emphasize distributed and collective intelligence, especially in systems integrating pervasive computing and communication technologies. This paper presents a preliminary study with the future goal of providing the necessary networking support for a scalable augmented reality (AR) IoT data visualization system for environmental intelligence. The system aims to apply edge computing and a distributed MQTT architecture to enable real-time data collection, processing, and visualization across diverse sensors and AR platforms. We evaluate the performance of a distributed MQTT broker system in a simulated multi-region IoT environment. Key performance metrics include latency, scalability, traffic overhead, resource utilization, and the impact of client locality. Comparisons with centralized setups highlight the benefits of distribution, such as resilience and load balancing, while also identifying areas for optimization. The findings provide actionable insights for designing a scalable and reliable IoT communication infrastructure.