A Distributed Computing Approach to IoT Data Processing: Edge, Fog, and Cloud Analytics Framework

Abstract

Background: An enormous amount of real-time data is produced by the spread of IoT devices in smart cities. Making decisions requires effective processing of this data. The amount, velocity, and variety of IoT data present difficulties for traditional centralised systems, resulting in significant latency and inefficient use of resources.
Objective: A hybrid Edge-Fog-Cloud architecture is suggested in this study to increase the latency, scalability, and processing efficiency of Internet of Things systems. In smart city applications, the objective is to guarantee precise data analytics, real-time decision-making, and efficient resource utilisation.
Methods: The design uses dynamic orchestration algorithms to assign jobs according to complexity and available resources, integrating the Edge, Fog, and Cloud computing layers. In order to compare with current techniques like SVM, NSGA-III, and MEC, performance measures including accuracy, efficiency, latency, and resource utilisation are assessed.
Results: The suggested method performs better than conventional methods with 90 ms latency, 94% efficiency, and 93% accuracy. It ensures optimal performance in real-time Internet of Things applications by achieving greater scalability and resource utilisation. It exhibits a notable decrease in processing time and power usage as compared to alternative approaches.
Conclusion: The Edge-Fog-Cloud framework improves IoT system performance, scalability, and energy efficiency. Large-scale, resource-constrained IoT systems, such as smart cities, can benefit from its ability to process and make decisions in real-time. It is an excellent option for upcoming IoT applications because to its exceptional efficiency.