Analisis Task Scheduling Menggunakan Optimized Particle Swarm Optimization (OPSO) dan Elite Opposition-Based Learning Pada Cloud Environment

Faizal, Wisnuyasha (2025) Analisis Task Scheduling Menggunakan Optimized Particle Swarm Optimization (OPSO) dan Elite Opposition-Based Learning Pada Cloud Environment. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5027211036-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (4MB) | Request a copy

Abstract

Efisiensi penjadwalan tugas dalam cloud computing adalah tantangan utama untuk mengoptimalkan sumber daya dan waktu eksekusi. Penelitian ini mengkaji penerapan Optimized Particle Swarm Optimization (OPSO) yang diintegrasikan dengan Elite Opposition-Based Learning (EOBL) untuk meningkatkan performa penjadwalan tugas di lingkungan cloud. OPSO memanfaatkan pendekatan Dynamic Adjustment of Parameters with Discrete Positioning (DAPDP) untuk memastikan keseimbangan antara eksplorasi dan eksploitasi dalam mencari solusi; sedangkan EOBL memperkaya ruang pencarian melalui solusi oposisi elit untuk menghindari jebakan local optimum. Implementasi dilakukan menggunakan CloudSim dengan tiga dataset: SDSC; Stratified Random; dan Simple Random. Hasil penelitian menunjukkan bahwa OPSO unggul pada dataset SDSC dengan makespan lebih baik sekitar 3,35%; throughput lebih tinggi sekitar 3,39%; dan resource utilization lebih tinggi sekitar 3,04%. OPSO juga sedikit unggul pada dataset Simple Random dengan makespan lebih baik sekitar 1,66%; throughput lebih tinggi sekitar 2,25%; resource utilization lebih tinggi sekitar 2,16%; dan total energy consumption lebih baik sekitar 1,03%. Sementara itu, OPSO EOBL menunjukkan performa terbaik pada dataset Stratified Random dengan makespan lebih baik sekitar 2,83%; throughput lebih tinggi sekitar 2,55%; dan resource utilization lebih tinggi sekitar 2,35%. Pada dataset SDSC, OPSO EOBL unggul dalam total energy consumption sekitar 0,40% lebih baik dan imbalance degree sekitar 4,27% lebih baik. OPSO dan OPSO EOBL memberikan solusi optimal terutama pada dataset yang lebih realistis dan mencerminkan kondisi dunia nyata.
================================================================================================================================
Task scheduling efficiency in cloud computing is a critical challenge for optimizing resource utilization and execution time. This study examines the implementation of Optimized Particle Swarm Optimization (OPSO) integrated with Elite Opposition-Based Learning (EOBL) to enhance task scheduling performance in cloud environments. OPSO employs the Dynamic Adjustment of Parameters with Discrete Positioning (DAPDP) approach to balance exploration and exploitation in finding solutions; while EOBL enriches the search space through elite opposition solutions to avoid local optima traps. The implementation was carried out using CloudSim with three datasets: SDSC; Stratified Random; and Simple Random. The results show that OPSO outperformed others on the SDSC dataset with a makespan approximately 3,35% better; throughput around 3,39% higher; and resource utilization about 3,04% higher. OPSO also achieved marginally better results on the Simple Random dataset; with a makespan approximately 1,66% better; throughput around 2,25% higher; resource utilization about 2,16% higher; and total energy consumption roughly 1,03% better. Meanwhile, OPSO EOBL demonstrated the best performance on the Stratified Random dataset; achieving a makespan approximately 2,83% better; throughput about 2,55% higher; and resource utilization roughly 2,35% higher. On the SDSC dataset, OPSO EOBL excelled in total energy consumption with an improvement of approximately 0,40% and in imbalance degree with an improvement of about 4,27%. Both OPSO and OPSO EOBL provided optimal solutions; particularly on datasets that are more realistic and reflective of real-world scenarios.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Cloud Computing, Elite Opposition-Based Learning, Optimized Particle Swarm Optimization, Task Scheduling.
Subjects:	T Technology > T Technology (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.6 Operations research--Mathematics. Goal programming T Technology > T Technology (General) > T57.62 Simulation T Technology > T Technology (General) > T57.84 Heuristic algorithms. T Technology > T Technology (General) > T58.8 Productivity. Efficiency
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology > 59201-(S1) Undergraduate Thesis
Depositing User:	Wisnuyasha Faizal
Date Deposited:	23 Jan 2025 08:40
Last Modified:	23 Jan 2025 08:40
URI:	http://repository.its.ac.id/id/eprint/116756

Actions (login required)

View Item