Comparative Analysis of MongoDB and InfluxDB for Time Series Data Management in IoT Environments: A Study on Performance, Scalability, and Concurrency
Piyush Tripathi, Mahdi H. Miraz, Snigdha Joshi
Abstract
The paper addresses the lack of comparative research on the performance and scalability of MongoDB and InfluxDB in managing high-concurrency workloads specific to Internet of Things (IoT) applications. Using a Python-based client script, we explore latency, resource usage, and scalability at diverse levels of concurrency. Experimental outcomes indicate that under high-concurrency IoT workloads, MongoDB outperforms InfluxDB in scalability, with latency falling to 0.003720 seconds at 50 concurrency, compared to InfluxDB’s 0.003374 seconds. InfluxDB shows better initial memory efficiency but struggles with CPU demand, spiking at 607.45% growth in utilization. MongoDB, despite higher memory use, remains stable in latency, showcasing better resource management with a lower CPU growth rate of 144.66%. These findings provide a framework for selecting appropriate database solutions in the IoT context.