Leveraging Community-based Approaches for Enhancing Resource Allocation in Fog Computing Environment

Abstract

Efficient resource allocation in fog computing environments is essential to address the increasing demand for high-performance and adaptable network services. Traditional methods lack granular differentiation based on traffic characteristics often resulting in suboptimal bandwidth utilization and elevated latency. To enhance network efficiency, this study applies a community-based resource allocation approach leveraging the Louvain algorithm to dynamically cluster network nodes with similar traffic demands. By forming communities based on bandwidth and latency needs, this approach enables a targeted resource distribution, aligning each community with optimized pathways that address specific requirements. The results indicate notable performance gains, including a 14% increase in bandwidth utilization affecting the download and a reduction in latency by an average of 23% for time-sensitive applications. These improvements highlight the effectiveness of the proposed approach in managing diverse network demands, improving data flow stability, and enhancing the overall performance of fog computing infrastructures. These findings underscore the potential for community-based resource allocation to support scalable, adaptable, and secure resource management, positioning it as a viable solution to meet the complex needs of IoT and other distributed network systems.