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
Hasbi Razzak
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 
Mekanisme Pembobotan Server Menggunakan Algoritme Fuzzy Pada Sistem Load Balancing di Software Defined Network Hasbi Razzak; Widhi Yahya; Mochammad Ali Fauzi
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 3 No 9 (2019): September 2019
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1093.804 KB)

Abstract

Load balancing is used to distribute workloads evenly to multiple servers. To achieve an optimal load balancing system must consider the workload received by the server based on the server's ability. As for what affects server performance includes resources such as CPU, memory and disk. If the workload received exceeds the server's capability, the server may be overloaded. To overcome this problem, we can utilize load balancing by implementing an algorithm that can take into account CPU, memory and disk variables. For this research, implementing a server weighting system uses a fuzzy algorithm on a load balancing system to distribute server workloads based on CPU, memory and disk variables. From the results and analysis for testing on this system, Fuzzy algorithms are able to distribute traffic based on the lowest server weights. Then a comparison with the Response time (RT) algorithm and the Least Connection (LC) algorithm with traffic distribution parameters, CPU and Memory Usage, Response Time and Throughput. When testing with a number of requests 360, Fuzzy algorithms charge more for distribution to servers the lowest server weight. Then in testing memory usage, the response time algorithm has the lowest memory usage on each server, server1 18.7%, server2 32.4% and server3 62.2%. Furthermore, testing with loads at the level of 360 RT algorithm has a high response time of 393 m/s. In testing throughput algorithm the response time has the largest throughput with a value of 93.6 KB.
Co-Authors M. Ali Fauzi Widhi Yahya