Analisis Kinerja Cloud Task Scheduling Menggunakan Algoritma Chaotic Squirrel Search dan Opposition-Based Learning

Septinaputri, Gracetriana Survinta (2025) Analisis Kinerja Cloud Task Scheduling Menggunakan Algoritma Chaotic Squirrel Search dan Opposition-Based Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025211199-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Cloud computing telah menjadi paradigma komputasi baru yang transformatif dan terus berkembang pesat sehingga dengan berbagai keunggulan yang ditawarkan. Namun, peningkatan permintaan layanan cloud mengakibatkan konsumsi energi yang masif dan tantangan dalam penjadwalan tugas yang efisien. Penelitian ini mengusulkan dua skenario simulasi penjadwalan tugas, yaitu skenario sistem penjadwalan tugas dengan menerapkan Chaotic Squirrel Search Algorithm (CSSA) dan kombinasi CSSA dengan Opposition-Based Learning (OBL). CSSA mengintegrasikan teori chaos, progressive search, dan jumping search untuk meningkatkan eksplorasi dan eksploitasi algoritma, sementara OBL diterapkan untuk menghindari konvergensi prematur. Simulasi dilakukan menggunakan framework CloudSim dengan membandingkan kinerja Genetic Algorithm (GA), SSA, CSSA, dan CSSA-OBL pada sepuluh parameter performa (makespan, total wait time, average start time, average finish time, average execution time, total scheduling length, throughput, resource utilization, imbalance degree, dan total energy consumption). Hasil eksperimen menunjukkan bahwa CSSA memberikan peningkatan kinerja dibandingkan GA, terbukti dengan peningkatan pada kesepuluh parameter evaluasi untuk simple random dataset, enam dari sepuluh parameter untuk stratified random dataset, dan sembilan dari sepuluh parameter untuk dataset SDSC. Sedangkan, implementasi OBL pada CSSA memberikan perbaikan pada beberapa parameter evaluasi di dataset simple random dan SDSC. Hasil penelitian membuktikan bahwa CSSA dapat diimplementasikan sebagai solusi untuk optimasi penjadwalan tugas cloud computing dengan OBL memberikan perbaikan tambahan terutama pada dataset dengan distribusi yang tidak merata.
=======================================================================================================================================
Cloud computing has become a new transformative computing paradigm that continues to develop rapidly, so with various advantages offered such as scalability, functionality, and cost using the "pay-as-you-go" principle, it results in continuously increasing adoption across various sectors. However, the increasing demand for cloud services leads to massive energy consumption and challenges in efficient task scheduling. This research proposes two task scheduling simulation scenarios: a task scheduling system scenario implementing Chaotic Squirrel Search Algorithm (CSSA) and a combination of CSSA with Opposition-Based Learning (OBL). CSSA integrates chaos theory, progressive search, and jumping search to improve algorithm exploration and exploitation, while OBL is applied to avoid premature convergence. Simulations were conducted using the CloudSim framework by comparing the performance of Genetic Algorithm (GA), SSA, CSSA, and hybrid CSSA-OBL across ten performance parameters. Experimental results show that CSSA provides performance improvements compared to GA, evidenced by improvements in all ten evaluation parameters for the simple random dataset, six out of ten parameters for the stratified random dataset, and nine out of ten parameters for the SDSC dataset. Meanwhile, the implementation of OBL on CSSA provides improvements in several evaluation parameters in the simple random and SDSC datasets. The research results prove that CSSA can be implemented as a solution for cloud computing task scheduling optimization, with OBL providing additional improvements especially on datasets with uneven distribution.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Chaotic Squirrel Search Algorithm, CloudSim, Komputasi Awan, Opposition Based Learning, Penjadwalan Tugas Chaotic Squirrel Search Algorithm, CloudSim, Cloud Computing, Opposition Based Learning, Task Scheduling
Subjects:	Q Science > QA Mathematics > QA76.585 Cloud computing. Mobile computing.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Gracetriana Survinta Septinaputri
Date Deposited:	29 Jul 2025 02:15
Last Modified:	29 Jul 2025 02:15
URI:	http://repository.its.ac.id/id/eprint/122469

Actions (login required)

View Item