Scalable Resource Management for Latency-Sensitive Applications in Edge Computing
Wahyu Adi Prijono, Primatar Kuswiradyo, Dipak Kumar Sahoo, Ali Mustofa, Sigit Kusmaryanto, Endah Budi Purnomowati · 2024
Edge computing is crucial for latency-sensitive Internet of Things (IoT) applications, providing reduced latency and enhanced computational efficiency. However, ensuring low latency while maintaining scalability presents significant challenges. This paper proposes a scalable resource management algorithm using heuristic methods to optimize resource allocation and workload distribution in edge computing environments. The algorithm dynamically adjusts to real-time demands, ensuring minimal latency and efficient utilization of computational resources. Extensive simulations were conducted to evaluate the performance of the proposed algorithm compared to the Greedy Algorithm and the First-Come, First-Served (FCFS) Algorithm. The results demonstrate that the proposed heuristic algorithm significantly outperforms the baseline algorithms in terms of number of packet losses, decreased average latency by up to 38%, and better resource utilization efficiency by up to 27%. The inclusion of buffers at edge nodes to handle non-real-time traffic allows the system to prioritize real-time traffic effectively, resulting in reduced packet drops and improved overall performance. These results highlight the algorithm's potential for improving the performance and scalability of edge computing for real-time IoT applications. Future research directions include exploring advanced optimization techniques, integrating energy-efficient resource management strategies, addressing security and privacy challenges, and conducting real-world deployment and validation of the proposed algorithm.