Enhanced Auto-scaling by using Dynamic scaling policy with Deep learning :LSTM

Ravikumaraiah, Mohith Srivathsa (2022) Enhanced Auto-scaling by using Dynamic scaling policy with Deep learning :LSTM. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (4MB) | Preview
Preview	PDF (Configuration manual) Download (8MB) | Preview

Abstract

Cloud computing is becoming the main pillar of strength for most of the new technologies because of its ability to scale and change. Auto-scaling systems are being used to make this kind of flexibility on demand possible. When it comes to AWS auto scaling it has the EC2 autoscaling policies. For ensuring the exact number of available EC2 instances correctly, EC2 auto scaling groups are used. this Autoscaling groups will also help the application to handle the incoming loads. In this paper, an auto-scaling system with deep learning technology is proposed. AutoScaling best steps are chosen based on what needs to be done. In this paper, a deep learning method of enhancing auto-scaling groups is implemented. In the initial stage, different deep learning methods are applied to the dataset and the results are evaluated. From the result, LSTM found to be the most appropriate method when compared to the BI-LSTM and Attention BI-LSTM methods. So, the Auto scaling groups are triggered by using this method.

Item Type:	Thesis (Masters)
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:	Tamara Malone
Date Deposited:	19 Dec 2022 15:03
Last Modified:	07 Mar 2023 17:28
URI:	https://norma.ncirl.ie/id/eprint/5997

Actions (login required)

View Item