Article Per Year (5 Year)
p-Index From 2020 - 2025
0
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Kevin Charlie
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Analisis Performa Load Balancing Pada Broker MQTT Menggunakan Algoritma Round Robin Kevin Charlie; Rakhmadhany Primananda; Mahendra Data
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 9 (2018): September 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (760.32 KB)

Abstract

MQTT is a communication protocol that requires a small resource and bandwidth. MQTT uses a broker in charge of connecting publishers and subscribers. MQTT brokers, especially Mosquitto, do not have mechanisms to overcome failures caused by CPU overload or device damage when only one broker is used. Thus, it takes load balancer and some brokers. In this research, a load balancer performance analysis using round robin algorithm is used to distribute MQTT brokers workload. The study was conducted on a single host using virtualization. There are 4 virtual devices, namely 3 brokers and 1 load balancer. Subscriber subscribes to the topic by accessing the load balancer address. Testing is done by using load balancer as differentiator variable to know impact given load balancer to broker. The result shows that load balancer with round robin algorithm is able to distribute the load evenly with the condition of all connected clients without any interruption with 13.56%, 13.68% and 15.12% CPU load for each broker with request of 400 that receive 10 publish message for each client. Then the system is able to reconnect to other brokers with a speed of 1.007 seconds when traffic is low and 1.005, 2.791, 4.593, and 2.005 seconds when traffic is high. Finally, the broker's message distribution takes about 0.362, 0.687, 0.891, 1.199, and 1.622 seconds for the number of publish messages of 25, 50, 75, 100, and 125.
Co-Authors Mahendra Data Rakhmadhany Primananda