Optimasi Load Balancing dan DNS pada CDN untuk Mempercepat Distribusi Konten menggunakan BIND dan HAProxy berbasis Linux

This research explores how to make content delivery faster using a Content Delivery Network (CDN) on Linux, focusing on optimizing DNS and load balancing with tools like BIND and HAProxy. It’s all about ensuring users get content quickly, no matter where they are. Set up three edge servers running HAProxy for load balancing and one BIND server for DNS, using a feature called geodns to direct users to the nearest server. We simulated users from different locations to mimic real-world use. Data was collected using tools like tcpdump for network traffic, Apache Bench for HTTP performance, and system monitoring tools to track CPU and memory usage. We measured latency (time to first byte), throughput (data speed), CPU usage, and error rates. Baseline: No optimizations, with latency at 244.2 ms, throughput at 100 MB/s, CPU usage at 70% with 15% variation, and 5% error rate. Geodns Only: Latency dropped to 111.2 ms, keeping throughput at 100 MB/s. Load Balancing Only: Throughput rose to 120 MB/s, CPU usage balanced to 60% with 5% variation, and error rate fell to 3%. Combined: Best results with latency at 107.2 ms, throughput at 120 MB/s, and error rate at 2%. An interesting detail is how geodns not only lowers latency but also helps balance server loads, which is especially useful in places like Indonesia with uneven user distribution. Combining both optimizations significantly boosts CDN performance, making it a great approach for content providers. It’s clear this can improve user experience, especially in areas with varied network setups.

Authors

Publication Details

Journal
KERNEL Jurnal Riset Inovasi Bidang Informatika dan Pendidikan Informatika
Published
2025-04-27
DOI
https://doi.org/10.31284/j.kernel.2024.v5i1.7590
Primary Topic
Environmental Engineering and Cultural Studies
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Optimasi Load Balancing dan DNS pada CDN untuk Mempercepat Distribusi Konten menggunakan BIND dan HAProxy berbasis Linux

Moh Noor Al-Azam, Cahyo Darujati, Novi Mustar Prihadi
KERNEL Jurnal Riset Inovasi Bidang Informatika dan Pendidikan Informatika
Environmental Engineering and Cultural Studies
article

Optimasi Load Balancing dan DNS pada CDN untuk Mempercepat Distribusi Konten menggunakan BIND dan HAProxy berbasis Linux

Moh Noor Al-Azam, Cahyo Darujati, Novi Mustar Prihadi
article en

Abstract

This research explores how to make content delivery faster using a Content Delivery Network (CDN) on Linux, focusing on optimizing DNS and load balancing with tools like BIND and HAProxy. It’s all about ensuring users get content quickly, no matter where they are. Set up three edge servers running HAProxy for load balancing and one BIND server for DNS, using a feature called geodns to direct users to the nearest server. We simulated users from different locations to mimic real-world use. Data was collected using tools like tcpdump for network traffic, Apache Bench for HTTP performance, and system monitoring tools to track CPU and memory usage. We measured latency (time to first byte), throughput (data speed), CPU usage, and error rates. Baseline: No optimizations, with latency at 244.2 ms, throughput at 100 MB/s, CPU usage at 70% with 15% variation, and 5% error rate. Geodns Only: Latency dropped to 111.2 ms, keeping throughput at 100 MB/s. Load Balancing Only: Throughput rose to 120 MB/s, CPU usage balanced to 60% with 5% variation, and error rate fell to 3%. Combined: Best results with latency at 107.2 ms, throughput at 120 MB/s, and error rate at 2%. An interesting detail is how geodns not only lowers latency but also helps balance server loads, which is especially useful in places like Indonesia with uneven user distribution. Combining both optimizations significantly boosts CDN performance, making it a great approach for content providers. It’s clear this can improve user experience, especially in areas with varied network setups.

KERNEL Jurnal Riset Inovasi Bidang Informatika dan Pendidikan InformatikaVol. 5(1)
Openalex Percentile: Top 88%
Environmental Engineering and Cultural Studies
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.