Implementation and Analysis of Clustering Techniques in Fog Computing Against Traditional Cloud Computing

The exponential expansion of the Internet of Things (IoT) has increased the demand for efficient, low-latency, and scalable computing solutions. Traditional cloud computing often fails to meet these requirements because of its high latency and network congestion. Fog computing, which decentralizes the computational resources to the network edge, is a promising alternative. This study investigates the implementation and analysis of dynamic clustering techniques-dynamic load-based clustering and dynamic location-based clustering-within fog computing environments, using iFogSim2 as the simulation platform. By comparing these techniques with traditional cloud computing architectures, the research highlights their impact on latency reduction and network usage optimization. Simulation results demonstrate the superiority of fog computing architectures with dynamic clustering in adapting to real-time IoT demands, minimizing delays, and improving overall system performance. The findings emphasize the role of clustering, particularly locationbased clustering, in advancing IoT-enabled applications and provide insights into designing robust, responsive, and efficient computing frameworks for modern IoT ecosystems.