Using step scaling policy with deep learning to enhance autoscaling

Paramasivam Malarkodi, Rajkumar (2024) Using step scaling policy with deep learning to enhance autoscaling. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (473kB) | Preview

Abstract

This study deals with Deep Learning models’ implementation through associative learningscaling policies in cloud computing. It aims to improve the effectiveness of automatedscaling approaches since conventional ones show constraints concerning dynamic workloadsand unusual traffic patterns. By selecting Deep Learning tools, such as Recurrent Neural Networks, and Long Short-Term Memory networks, it also targets ensuring the correctness of resource allocation. The modeling part is followed by intensive experiments, including data pre-processing and selection of features, to check the applicability of the chosen techniques. The utilization of Deep Learning methods resulted in higher scalability and the ability to respond to predicted scaling opportunities. The part devoted to practical concerns, such as data privacy and the possibility of using the proposed models, shows that ethical considerations are relevant. Therefore, the findings of the presented study can be useful in terms of cloud resource management and allowing the way for the further development of scaling solutions.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Kumar Sharma, Jitendra UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	04 Jul 2025 10:02
Last Modified:	04 Jul 2025 10:02
URI:	https://norma.ncirl.ie/id/eprint/8046

Actions (login required)

View Item