NORMA eResearch @NCI Library

A Novel Failure Recovery Technique On AWS Spot Instances With Optimal Cloud Cost

Ravichandran, Sarath (2020) A Novel Failure Recovery Technique On AWS Spot Instances With Optimal Cloud Cost. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
Preview
PDF (Master of Science)
Download (2MB) | Preview
[thumbnail of Configuration manual]
Preview
PDF (Configuration manual)
Download (4MB) | Preview

Abstract

The main purpose of this research is to guarantee the reliability of the service that is provided by the AWS(Amazon Web Services)spot instance. A novel spot instance failure recovery life-cycle is proposed to assure the reliability with the minimal cloud cost. This proposal comprises two main modules in it namely forecast and spot instance failure recovery. A unique spot instance failure recovery algorithm is also introduced as a part of this study. Data loss due to spot instance termination is overcome by the suggested algorithm. A web application is built in order to maintain the life cycle of the spot instance. The proposed algorithm would initiate the spot instance AMI creation when the forecasted spot price is higher than the user bid price. This AMI would be used to initiate an on-demand instance if the spot instance is terminated. Thus the data loss due to the spot instance termination by the vendor due to lower bid-price is resolved. Once the on-demand instance is created then the application would continuously monitor the user bid price and current spot price. Once the current spot price is equal to the user bid price then the application would generate an AMI for the running on-demand instance and this AMI would be used to create an equivalent spot instance. Thus the cost spent on the cloud is significantly decreased. This proposal is implemented and evaluated on the AWS cloud platform. The evaluation results have concluded that the business continuity is achieved even after the termination of spot instance.

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
Divisions: School of Computing > Master of Science in Cloud Computing
Depositing User: Dan English
Date Deposited: 28 Jan 2021 17:17
Last Modified: 28 Jan 2021 17:17
URI: https://norma.ncirl.ie/id/eprint/4547

Actions (login required)

View Item View Item