LOAD BALANCING ALGORITHM FOR FOG COMPUTING

Abstract

This research paper presents a comparative analysis of cloud and fog computing environments based on key performance parameters, including response time, processing time, virtual machine (VM) cost, and data transfer cost. The study evaluates two resource allocation policies—Closest Fog Node and Optimized Response Time—to determine their impact on system efficiency and user experience. Simulation results reveal that the fog computing environment significantly outperforms the traditional cloud setup, achieving an average response time of approximately 124.8 ms, compared to 699.8 ms in the cloud. Although the VM cost in the fog setup is slightly higher, the substantial reduction in latency and enhanced responsiveness make fog computing a more suitable architecture for real-time and latency- critical applications. The findings further indicate that optimization-based policies provide marginal improvements in performance compared to simple proximity-based allocation. Overall, the study highlights the effectiveness of fog computing in delivering faster, more reliable, and location-aware services, and suggests that integrating intelligent resource allocation models can further improve performance and energy efficiency in future distributed computing environments.