Analisis Penyeimbangan Beban Kerja di Cloud Environment Menggunakan Flower Pollination Algorithm (FPA)

Setyowati, Della (2025) Analisis Penyeimbangan Beban Kerja di Cloud Environment Menggunakan Flower Pollination Algorithm (FPA). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5027211044_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 2025. Download (5MB) | Request a copy

Abstract

Seiring meningkatnya penggunaan cloud computing dalam penyediaan layanan komputasi yang fleksibel dan efisien, tantangan dalam mendistribusikan beban kerja secara optimal semakin penting. Penelitian ini mengusulkan dan mengimplementasikan algoritma Flower Pollination Algorithm (FPA), sebuah metode metaheuristik yang terinspirasi dari proses penyerbukan bunga di alam, untuk menyelesaikan masalah load balancing di lingkungan cloud. Dengan membandingkan performa FPA dengan algoritma pembanding seperti Genetic Algorithm (GA), Particle Swarm Optimization (PSO), dan Ant Colony Optimization (ACO), baik melalui simulasi menggunakan CloudSim maupun pada implementasi nyata berbasis container Docker. Dataset yang digunakan mencakup Simple Random, Stratified Random, dan SDSC Blue Horizon Log. Dan evaluasi dilakukan menggunakan metrik performa seperti start time, wait time, execution time, finish time, makespan, dan imbalance degree.
Hasil penelitian menunjukkan FPA menghasilkan distribusi beban kerja yang lebih merata, dengan imbalance degree 0,025355 (Simple Random), 0,177169 (Random Stratified),
0,11874 (SDSC), dan 0,127 (real environment), mencerminkan pemerataan beban 50–70% lebih baik dibandingkan algoritma pembanding. FPA juga mengurangi average wait time pada
Random Stratified (1.929,9977) dan SDSC (1.956,92), tetapi mencatat makespan lebih tinggi (1.577.230,806) dibandingkan dengan GA (695.340,4057) pada Random Stratified Dataset.
Dengan kemampuan adaptifnya, FPA terbukti sebagai solusi kompetitif untuk load balancing pada sistem cloud modern.
========================================================================================================================
====================
As the adoption of cloud computing continues to grow in delivering flexible and efficient computing services, the challenge of optimally distributing workloads has become increasingly critical. This study proposes and implements the Flower Pollination Algorithm (FPA), a metaheuristic method inspired by the natural process of flower pollination, to address the issue of load balancing in cloud environments.
The performance of FPA is compared with other algorithms such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO), through both simulation using CloudSim and real-world implementation based on Docker containers. The datasets used include Simple Random, Stratified Random, and SDSC Blue Horizon Log. The evaluation is conducted using performance metrics such as start time, wait time, execution time, finish time, makespan, and imbalance degree.The results show that FPA achieves a more balanced workload distribution, with imbalance degrees of 0.025355 (Simple Random), 0.177169 (Stratified Random), 0.11874 (SDSC), and 0.127 (real environment), reflecting a 50–70% better load distribution compared to the other algorithms. FPA also reduces the average wait time on the Stratified Random dataset (1,929.9977) and SDSC dataset (1,956.92), although it records a higher makespan (1,577,230.806) compared to GA (695,340.4057) on the Stratified Random dataset. With its adaptive capabilities, FPA proves to be a competitive solution for load balancing in modern cloud systems.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Flower Pollination Algorithm, Load balancing, Cloud computing, Metaheuristik, CloudSim, Docker.
Subjects:	Q Science > QA Mathematics > QA76.585 Cloud computing. Mobile computing.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology > 59201-(S1) Undergraduate Thesis
Depositing User:	Della Setyowati
Date Deposited:	01 Jul 2025 03:09
Last Modified:	01 Jul 2025 03:09
URI:	http://repository.its.ac.id/id/eprint/119291

Actions (login required)

View Item