MSCEIS 2019
Submission Management System
Main Site
Submission Guide
Register
Login
Participant List
Abstract List
Paper List
Access Mode
Contact
:: Abstract ::

<< back

COMPARISON OF ROUND ROBIN AND LEAST CONNECTION ALGORITHM ON HAPROXY FOR LOAD BALANCING WEB SERVER
Arip Solehudin(a*), Purwantoro(a), Fahmi Yuda Fauzi(a), Munir(,b)

a) Faculty of Computer Science, Universitas Singaperbangsa Karawang, Karawang 41361, Indonesia
*arip.solehudin[at]staff.unsika.ac.id, purwantoro.masbro[at]staff.unsika.ac.id, fahmi.yuda15047[at]student.unsika.ac.id
b)Department of Computer Science Education, Universitas Pendidikan Indonesia
munir[at]upi.edu


Abstract

Technology development is currently experiencing a fairly rapid increase, Therefore at the high school in Karawang intends to hold a grade promotion test using the Computer Based Test (CBT). However, in its implementation there are problems one of which is the server load. This is caused by user requests that exceed the servers capabilities. To overcome these problems, we use web server load balancing using haproxy. But the use of web server load balancing, often experience data inconsistencies per server. Therefore, database replication is also used that can guarantee data consistency in web server load balancing. This research uses descriptive approach with PPDIOO network system design method. There are 2 load balancing algorithms compared, namely round robin and least connection. Based on the httperf test results, throughput parameters of the two algorithms obtain the same average value of 99.5 Kb / second. In testing the response time parameter the least connection algorithm is superior at 6.9 ms compared to round robin which has an average value of 7.2 ms, but in testing the CPU Utilization parameter the round robin algorithm is superior at 23.7% compared to the least connection algorithm at 24.3%.

Keywords: Computer Based Test, web server load balancing, database replication

Topic: Computer Science

Plain Format | Corresponding Author (Arip Solehudin)

PermaLink

MSCEIS 2019 - Submission Management System

Powered By Konfrenzi 1.832K-Build2 © 2025 All Rights Reserved