Reliable Scalable Storage: Solution for cloud native applications

Thumbnail Image

Files

master_Skipwith_Grey_2024.pdf (537.2 KB)

URL

Journal Title

Journal ISSN

Volume Title

Perustieteiden korkeakoulu | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2024-01-22

Department

Major/Subject

Cloud Computing and Services

Mcode

SCI3081

Degree programme

Master's Programme in ICT Innovation

Language

en

Pages

75

Series

Abstract

In an inter-connected world where the role of data is primordial and more important than ever, the traditional monolithic storage systems encounter some issues to provide the necessary accessibility and reliability of data. Therefore, a reevaluation of storage methodologies is needed, particularly in addressing the growing need for a scalable, reliable, and cost-effective storage solution. This thesis starts by highlighting the limitations of the monolithic storage system and the importance of data in Ericsson. This thesis aims to provide a literature evaluation of different software-defined storage (SDS) technologies to understand their capabilities and suitability to replace the current monolithic storage system at Ericsson and overcome its issues. We explore a range of SDS technologies, Ceph, GlusterFS and HDFS and, based on existing literature, surveyed their performances and mechanisms in terms of performance, resilience and replication. Our analysis is structured around key research questions aimed at identifying the mechanisms for a resilient storage, evaluating the technologies for redundancy and cost-effectiveness, and highlighting the strategies making a technology reliable. We have concluded that while all investigated technologies offer significant potential for modern storage needs and meet our requirements, in our case, GlusterFS would be the best fit. This technology has displayed great performances and native features to ensure the reliability of the data while decreasing the metadata overhead encountered in the other technologies. However, this outcome is a theoretical evaluation of the technologies and further testing would be necessary to fully comprehend the behaviour of each technology in our environment. Therefore, this thesis contributes to the understanding of SDS technologies by comparing them to real-world requirements thus, making the first step in the decision making process of transitioning from a traditional storage to a SDS technology.

Description

Supervisor

Truong, Linh

Thesis advisor

Pusca-Horeanga, Alexandru

Keywords

software-defined storage, storage, replication, scalability, downtime, recovery

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-202401282074

Collections

[dipl] Perustieteiden korkeakoulu / SCI