Performance Evaluation of Edge Computing for Latency-Sensitive Applications

Abstract

Edge computing has emerged as a critical paradigm for supporting latency-sensitive applications such as autonomous driving, real-time health monitoring, augmented reality (AR), and industrial automation. By decentralizing computational resources closer to end devices, edge computing minimizes round-trip latency and bandwidth usage compared to centralized cloud architectures. This study evaluates the performance of edge computing frameworks in handling low-latency demands under varying workloads and network conditions. The paper explores key performance metrics—latency, throughput, reliability, and energy efficiency—and presents a comparative analysis with cloud-centric approaches. Findings reveal that edge architectures offer significant latency reductions, averaging 40–60%, while maintaining comparable accuracy and stability. Moreover, intelligent task offloading and load-balancing algorithms further enhance overall system performance. The evaluation underscores edge computing’s pivotal role in enabling next-generation Internet of Things (IoT) ecosystems and time-critical applications.