YR, Muh Hilmy Thoriq (2023) Comparison Study of Task Distribution Strategy Based on Load Balanced Markov Process Modelling, Probabilistic Load Balancing Ant Colony, and Simulated Annealing. Other thesis, Institut Teknologi Sepuluh Nopember Surabaya.
![[thumbnail of 05111942000012_Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
05111942000012_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 September 2025. Download (2MB) | Request a copy |
Abstract
Dalam pengaturan cloud computing, tugas-tugas didelegasikan di antara banyak virtual machines (VMs), masing-masing ditandai oleh panjang tugas, waktu peluncuran, dan durasi pemrosesan yang unik. Untuk mengoptimalkan penggunaan dan meningkatkan kinerja sistem, sangat penting untuk menyebar beban kerja secara merata di antara VMs melalui load balancing dan kita membutuhkan strategi distribusi tugas terbaik. Beberapa strategi distribusi tugas yang umum seperti Round Robin, First-Come First�Serve (FCFS), Shortest Job First (SJF) dan lain-lain, mungkin tidak mendistribusikan tugas dengan mempertimbangkan keadilan. Memilih strategi distribusi tugas yang tepat sangat
penting untuk menjaga sistem tetap seimbang. Dalam makalah ini, kami memberikan perbandingan antara strategi distribusi tugas yang mencoba load balancing, yaitu pemodelan Markov process yang seimbang yang akan kami fokuskan, Probabilistic Load Balancing Ant Colony (PLAC), dan Simulated Annealing untuk memutuskan strategi distribusi tugas mana yang berhasil meningkatkan parameter seperti makespan, resource utilization, dan degree of imbalance serta menjadi pilihan terbaik. Berdasarkan hasil pengujian yang dilakukan dapat disimpulkan bahwa pemodelan Markov process modelling mengalahkan algoritma pembanding lainnya dalam sisi performa.
===================================================================================================================================
In cloud computing settings, tasks are delegated among numerous virtual machines (VMs), each characterized by its unique task length, launch time, and processing duration. To optimize usage and boost system performance, it's crucial to evenly spread the workload among the VMs through load balancing, and we need the best task distribution strategy. Some of the common task distribution strategies, such as Round Robin, First-Come First�Serve (FCFS), and Shortest Job First (SJF), may not distribute tasks with fairness in mind. In this paper, we provide a comparison between task distribution strategies that involve load balancing. Specifically, we focus on load-balanced Markov process modeling, Probabilistic Load Balancing Ant Colony (PLAC), and Simulated Annealing. Our goal is to determine which task distribution strategy performs well in terms of improving parameters like makespan, resource utilization, and degree of imbalance. By examining these strategies, based on our experiment result we may conclude that Load Balanced Markov Process Modelling is outperforms the other algorithms.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Load Balancing, Cloud Computing, Task Distribution, Markov Modelling, Expected utilization, Expected processing capacity |
| 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: | Muh Hilmy Thoriq YR |
| Date Deposited: | 09 Oct 2023 01:07 |
| Last Modified: | 09 Oct 2023 01:07 |
| URI: | http://repository.its.ac.id/id/eprint/103191 |
Actions (login required)
 |
View Item |