Optimal Distributed MQTT Broker and Services Placement for SDN-Edge Based Smart City Architecture

Area of Science:

• Computer Science
• Internet of Things (IoT)
• Distributed Systems

Background:

• Smart cities leverage IoT infrastructure for resource management, utilizing protocols like Message Queuing Telemetry Transport (MQTT).
• Edge computing offers distributed resources closer to data sources, but centralized MQTT brokers pose challenges like latency and bottlenecks in such environments.

Purpose of the Study:

• To propose and evaluate an optimized distributed MQTT broker architecture for smart city edge computing environments.
• To address the limitations of centralized MQTT brokers in distributed edge resource scenarios.

Main Methods:

• Developed a novel distributed MQTT broker architecture tailored for edge resources.
• Formulated an integer non-linear program to optimize container placement for efficient resource utilization.
• Conducted extensive simulations comparing the proposed architecture against existing distributed MQTT middleware with greedy and random placement strategies.

Main Results:

• The proposed distributed MQTT broker architecture significantly reduced network traffic and data delivery latency.
• Demonstrated superior performance in lowering deployment failure ratio, power consumption, network usage, and synchronization overhead compared to existing methods.
• Optimized container placement effectively managed edge computing resources.

Conclusions:

• The optimized distributed MQTT broker architecture is highly effective for smart city IoT applications at the edge.
• This approach enhances the efficiency, reliability, and performance of data management in distributed IoT systems.