Optimasi Task Scheduling Menggunakan Artificial Bee Colony (ABC) dan Elite Opposition-based Learning (EOBL) Pada Cloud Environment

Uzlahwasata, Dzakirozaan (2025) Optimasi Task Scheduling Menggunakan Artificial Bee Colony (ABC) dan Elite Opposition-based Learning (EOBL) Pada Cloud Environment. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5027211066-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Penjadwalan tugas (task scheduling) yang efisien pada lingkungan cloud computing merupakan tantangan krusial untuk memaksimalkan penggunaan sumber daya. Algoritma Artificial Bee Colony (ABC) yang umum digunakan sering kali mengalami kelemahan dalam hal pemanfaatan sumber daya dan cenderung terjebak pada solusi lokal. Untuk mengatasi masalah ini, penelitian ini mengusulkan optimasi melalui kombinasi algoritma ABC dengan pendekatan Elite Opposition-Based Learning (EOBL). Metode ini diimplementasikan dengan strategi oposisi berdasarkan solusi elite guna memperluas ruang pencarian dan menghindari konvergensi prematur. Hasil pengujian menunjukkan bahwa pendekatan ABC-EOBL secara signifikan meningkatkan keseimbangan alokasi sumber daya, terbukti dari performa parameter imbalance degree yang 18,40% lebih efisien dibandingkan algoritma pembanding. Meskipun unggul dalam parameter imbalance degree, penerapan EOBL menyebabkan penurunan kinerja pada metrik lain seperti makespan, average start time, average finish time, average execution time, average waiting time, scheduling length, throughput, resource utilization, dan energy consumption, akibat meningkatnya kompleksitas pencarian. Kesimpulannya, metode ABC-EOBL menawarkan keuntungan signifikan untuk distribusi sumber daya yang seimbang, namun penggunaannya harus mempertimbangkan adanya trade-off dengan efisiensi waktu eksekusi.

======================================================================================================================================

Efficient task scheduling in cloud computing environments is a crucial challenge for maximizing resource utilization. The commonly used Artificial Bee Colony (ABC) algorithm often suffers from weaknesses in resource utilization and a tendency to get trapped in local optima. To address this issue, this research proposes an optimization by combining the ABC algorithm with the Elite Opposition-Based Learning (EOBL) approach. This method is implemented using an opposition strategy based on elite solutions to expand the search space and avoid premature convergence. Experimental results show that the ABC-EOBL approach significantly improves the balance of resource allocation, evidenced by an 18.40% more efficient performance in the imbalance degree parameter compared to benchmark algorithms. Despite its superiority in the imbalance degree parameter, the implementation of EOBL leads to a decrease in performance across other metrics such as makespan, average start time, average finish time, average execution time, average waiting time, scheduling length, throughput, resource utilization, and energy consumption, due to increased search complexity. In conclusion, the ABC-EOBL method offers a significant advantage for balanced resource distribution, but its application must consider the trade-off with execution time efficiency.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Penjadwalan Tugas, Komputasi Awan, Artificial Bee Colony, Elite Opposition-based Learning.
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:	Dzakirozaan Uzlahwasata
Date Deposited:	29 Jul 2025 02:37
Last Modified:	29 Jul 2025 02:37
URI:	http://repository.its.ac.id/id/eprint/122474

Actions (login required)

View Item