Optimization of Web Server Load Balancing Using HAProxy with the Weighted Round Robin Algorithm
Zahrina Amalia, Sampurna Dadi Riskiono · Jurnal Teknologi Informasi dan Pendidikan · 2025
The advancement of information and communication technology (ICT) in Indonesia has driven the widespread use of web-based services across various sectors, including education, public services, and digital administration. However, the growing traffic demand is often not supported by adequate server infrastructure, leading to issues such as slow access and system failures. These problems are primarily caused by unbalanced server workloads, especially when relying on a single server or lacking effective load distribution strategies. Load balancing technology is therefore essential to ensure stable and efficient web service performance. This study explores the implementation of the Weighted Round Robin (WRR) algorithm using HAProxy software to distribute loads proportionally based on each server's capacity. The research employs a quantitative experimental method by building a test environment consisting of one HAProxy server and multiple backend servers with different specifications. Traffic simulations are conducted using Apache JMeter to evaluate system performance based on technical metrics such as response time and request success rate. The experimental results show that the implementation of WRR with HAProxy reduced the average response time by 35.7% and improved the request success rate to 99.2%, compared to a baseline scenario without load balancing. These findings demonstrate that WRR-based load balancing significantly enhances the reliability and efficiency of web services, offering practical solutions to support Indonesia’s digital transformation efforts amid limited infrastructure resources.