Implementasi Load Balancing Pada Cloud Computing Dengan Algoritme Improved Weighted Least Connection

Pujangga, Muhammad Denny Chrisna (2021) Implementasi Load Balancing Pada Cloud Computing Dengan Algoritme Improved Weighted Least Connection. Sarjana thesis, Universitas Brawijaya.

Abstract

Dewasa ini, Cloud computing banyak diminati karena kemampuannya dalam menyelesaikan berbagai masalah dalam dunia komputasi, salah satunya permasalahan sumber daya komputasi yang terbatas. Cloud computing menyediakan fasilitas sumber daya terdistribusi yang dapat diatur sesuai kebutuhan dan diakses dari manapun dan kapanpun. Namun, dengan banyaknya sumber daya terdistribusi pada sistem cloud, dibutuhkan sebuah mekanisme pendistribusian beban kerja untuk menjaga kinerja sistem cloud agar tetap stabil, yaitu load balancing. Load balancing merupakan salah satu mekanisme pendistribusian beban kerja yang banyak digunakan pada sistem terdistribusi. Salah satu algoritme load balancing yang banyak digunakan adalah Algoritme Weighted Least Connection. Algoritme Weighted Least Connection memiliki permasalahan ketika server baru ditambahkan ke sistem cluster server ketika sedang terdapat banyak user request, algoritme Weighted Least Connection akan mendistribusikan beban kerja secara intensif ke server baru, sehingga menyebabkan ketidakmerataan distribusi beban kerja server. Untuk mengatasi permasalahan ini, dilakukan pengembangan pada algoritme Weighted Least Connection, yaitu dengan menggabungkan dua algoritme, yaitu algoritme Server Selection dan algoritme Overload Reduce menjadi satu algoritme, yaitu algoritme Improved Weighted Least Connection. Penelitian ini menerapkan algoritme Improved Weighted Least Connection pada arsitektur cloud menggunakan simulator CloudSim Plus. Hasil dari penelitian ini menunjukkan bahwa algoritme Improved Weighted Least Connection terbukti dapat menyeimbangkan beban kerja semua server, terutama server baru dengan tidak memberikan beban berlebih secara terus menerus ke server baru dengan rata-rata 4,75 task per-10 detik dan rata-rata CPU usage 28,02%, dibandingkan dengan server lama dengan kapasitas sumber daya yang sama yaitu rata-rata 14,07 task per-10 detik dan rata-rata CPU usage 70,45%.

English Abstract

Cloud computing is in great demand nowadays because of its ability to solve various problems in the world of computing, one of which is the limited computing resources. Cloud computing provides distributed resource that can be configured on demand and accessed from anywhere and anytime. However, with the large number of distributed resources on a cloud system, a workload distribution mechanism is needed to keep the performance of the cloud system stable, namely load balancing. Load balancing is a workload distribution mechanism that is widely used in distributed systems. One of the most widely used load balancing algorithms is the Weighted Least Connection Algorithm. The Weighted Least Connection algorithm has a problem that when a new server is added to the server cluster system when there are many user requests, the Weighted Least Connection algorithm will distribute the workload intensively to the new server, causing an uneven distribution of server workloads. To overcome this problem, the Weighted Least Connection algorithm was developed by combining two algorithms, the Server Selection algorithm and the Overload Reduce algorithm into one algorithm, namely the Improved Weighted Least Connection algorithm. This study applies the Improved Weighted Least Connection algorithm on the cloud architecture using the CloudSim Plus simulator. The results of this study indicate that the Improved Weighted Least Connection algorithm is proven to be able to balance the workload of all servers, especially new servers by not continuously overloading new servers with an average of 4.75 tasks per 10 seconds and an average CPU usage 28.02%, compared to the old server with the same resource capacity, which has an average of 14.07 tasks per 10 seconds and an average CPU usage of 70.45%.

Other obstract

-

Item Type:	Thesis (Sarjana)
Identification Number:	0521015
Uncontrolled Keywords:	server, Cloud computing, load balancing, weighted least connection, cloudsim plus, server, cloud computing, load balancing, weighted least connection, cloudsim plus
Subjects:	000 Computer science, information and general works > 004 Computer science
Divisions:	Fakultas Ilmu Komputer > Teknik Informatika
Depositing User:	Unnamed user with username santoso
Date Deposited:	21 Oct 2021 04:21
Last Modified:	30 Sep 2024 06:40
URI:	http://repository.ub.ac.id/id/eprint/184492

Text -M. Denny Chrisna Pujangga.pdf Download (2MB)

Actions (login required)

View Item