Choosing a message broker for Internet of things based on load testing

This article describes load testing common IoT message brokers under various configurations. The list of message brokers is represented by Mosquitto, EMQX, Active MQ, and Aedes. The NBomber library was chosen as the main tool for load testing. Based on this library, scenarios for stress testing brokers were developed. Each scenario was applied to the corresponding number of connections, message size, and QoS level. A custom client (called "message listener") was developed to handle errors. The collected data included the number of requests processed per second, the latency of sending messages, the number of errors, and the CPU utilization. Based on the obtained data, a normalized broker rating was proposed and applied. The best message broker can be selected by calculating the normalized broker score for each load scenario and merging all the calculated results into a pivot table. The best message broker is selected from the result table.

Authors

Publication Details

Journal
Dynamics of Complex Systems - XXI century
Published
2025-01-01
DOI
https://doi.org/10.18127/j19997493-202503-04
Primary Topic
IoT-based Smart Home Systems
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Choosing a message broker for Internet of things based on load testing

Robert M. Klassen, Artem Solopiy
Dynamics of Complex Systems - XXI century
IoT-based Smart Home Systems
article

Choosing a message broker for Internet of things based on load testing

Robert M. Klassen, Artem Solopiy
article en

Abstract

This article describes load testing common IoT message brokers under various configurations. The list of message brokers is represented by Mosquitto, EMQX, Active MQ, and Aedes. The NBomber library was chosen as the main tool for load testing. Based on this library, scenarios for stress testing brokers were developed. Each scenario was applied to the corresponding number of connections, message size, and QoS level. A custom client (called "message listener") was developed to handle errors. The collected data included the number of requests processed per second, the latency of sending messages, the number of errors, and the CPU utilization. Based on the obtained data, a normalized broker rating was proposed and applied. The best message broker can be selected by calculating the normalized broker score for each load scenario and merging all the calculated results into a pivot table. The best message broker is selected from the result table.

Dynamics of Complex Systems - XXI centuryVol. 19(3)
Openalex Percentile: Top 76%
IoT-based Smart Home Systems
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Choosing a message broker for Internet of things based on load testing — Robert M. Klassen, Artem Solopiy · Dynamics of Complex Systems - XXI century (2025) | TGRS Research Map | TGRS