Cloud & IoT
What I Learned From Building a Fog Computing IoT System
For my dissertation project, I built a real-world IoT system to compare cloud-centric and fog-assisted data pipelines.
The system used ESP32 devices, a Raspberry Pi 5 fog node, MQTT messaging, SQLite logging, and Google Cloud infrastructure.
The main aim was to investigate how adding a fog node influenced end-to-end latency, data handling, and system behaviour under different conditions.
One of the biggest lessons was that fog computing is context dependent. In some environments, adding another processing layer can increase delay. In others, fog nodes can support local processing, filtering, resilience, and reduced cloud dependency.
This project strengthened my understanding of distributed systems, networking, IoT messaging, performance testing, and practical troubleshooting.