Performance Analysis of Edge and Cloud Computing Architectures for Real-Time IoT Irrigation Systems

Abstract

Edge computing has emerged as a promising solution for reducing communication latency in Internet of Things (IoT) applications that require real-time decision-making. This paper presents a comparative performance analysis of edge-and cloud-based computing architectures for a real-time IT irrigation system. A prototype consisting of a Raspberry Pi 3 Model B, a capacitive soil moisture sensor, and the Favoriot cloud platform was developed to evaluate the impact of processing location on system responsiveness. Both architectures were implemented using identical hardware, communication protocols, and operating conditions to ensure a fair comparison. Experimental results from five repeated trials show that the edge architecture achieved an average response time of 0.477 s, whereas the cloud architecture required 2.195 s, representing a 78.3% reduction in response time and an approximately 4.6-fold improvement in responsiveness. These findings demonstrate the effectiveness of edge computing for time-sensitive agricultural IoT applications and provide a controlled experimental comparison of edge and cloud architectures, offering a practical benchmark for future smart irrigation research.