Análise Comparativa de Desempenho do Raspberry Pi Zero W e Zero 2W para Computação de Borda em Cenários de IoT

Abstract

Edge computing has emerged as a promising solution to address the challenges of latency, bandwidth limitations, and real-time processing in Internet of Things (IoT) applications. This study presents a comparative performance analysis of the Raspberry Pi Zero W and Zero 2W in an edge computing scenario, focusing on key metrics such as CPU utilization, memory usage, and network traffic. The experimental setup emulates an IoT environment, where temperature and pressure sensors transmit data to the Raspberry Pi devices for local processing at the network edge. The results indicate that while both models are capable of handling the workload, the Zero 2W demonstrates significantly better CPU performance, maintaining 53.26% idle time compared to only 20.20% for the Zero W under similar loads. In contrast, the Zero W offers a more cost-effective solution with acceptable performance for applications that require lower data volumes, less intensive processing, or have more flexible latency requirements. This analysis supports the design of efficient and scalable edge computing architectures across various IoT domains, providing guidelines for selecting the most appropriate hardware based on specific application requirements.