PREDIKSI SISTEM AUTO-SCALING PADA CLOUD PRIVATE MENGGUNAKAN METODE AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA)

SINURAT, ANDREAS BONA FAJAR and Heryanto, Ahmad (2025) PREDIKSI SISTEM AUTO-SCALING PADA CLOUD PRIVATE MENGGUNAKAN METODE AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA). Undergraduate thesis, Sriwijaya University.

Preview	Image RAMA_56201_09011282126115_Cover.jpg - Cover Image Available under License Creative Commons Public Domain Dedication. Download (161kB) | Preview
	Text RAMA_56201_09011282126115.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_56201_09011282126115_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_56201_09011282126115_0022018703_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (850kB)
	Text RAMA_56201_09011282126115_0022018703_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (399kB) | Request a copy
	Text RAMA_56201_09011282126115_0022018703_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011282126115_0022018703_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (538kB) | Request a copy
	Text RAMA_56201_09011282126115_0022018703_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (160kB) | Request a copy
	Text RAMA_56201_09011282126115_0022018703_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (218kB) | Request a copy
	Text RAMA_56201_09011282126115_0022018703_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (403kB) | Request a copy

Abstract

Cloud computing offers flexibility in managing IT resources, particularly through auto-scaling mechanisms that automatically adjust capacity based on workload. In private cloud environments, auto-scaling plays a crucial role in maintaining service availability while minimizing resource waste. However, its effectiveness relies heavily on the system’s ability to accurately predict future workloads. This study aims to evaluate three time series forecasting models—ARIMA, Holt’s Linear Trend, and Simple Exponential Smoothing (SES)—in predicting CPU usage, RAM usage, and disk activity in a virtualized private cloud system. Historical resource data were analyzed to identify usage patterns. Results indicate that CPU and disk activity data exhibit fluctuating and non-stationary behavior, while RAM usage is relatively stable. The models were evaluated using RMSE, MAE, and MAPE metrics. Evaluation results show that the ARIMA model consistently yields the lowest prediction errors across all datasets. With its higher accuracy compared to the other models, ARIMA is recommended as the primary model to support efficient, responsive, and adaptive auto-scaling in private cloud environments. It effectively captures trend and seasonal patterns in data, enabling better resource allocation, reducing overprovisioning, and avoiding underutilization.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Cloud Computing, Auto-Scaling, Prediksi Beban Kerja, ARIMA, Time Series
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Andreas Bona Fajar Sinurat
Date Deposited:	25 Aug 2025 06:22
Last Modified:	25 Aug 2025 07:02
URI:	http://repository.unsri.ac.id/id/eprint/183163

Actions (login required)

View Item