On Improved Performance and Secure Data Management in Cloud Storage Services

Identifier:  TDX:4386

Handle: https://hdl.handle.net/20.500.11797/TDX4386

Authors:  Saiz Laudo, Raul

Abstract:
This thesis delves into two crucial aspects of digital storage services: personal cloud storage and cloud object storage. The first focus is on personal cloud storage services, specifically targeting the efficiency of file synchronization protocols. These protocols are essential for timely reflecting the changes made in a user's local file system in the cloud. However, poorly designed protocols can lead to excessive network traffic, resulting in significant financial losses and performance issues for both service providers and users. To address this, the thesis introduces the RSD algorithm that aims to strategically delay the sending of modifications in the synchronization process with the cloud. The second part of the thesis addresses limitations in data protection tools for cloud-based object storage services. It proposes EGEON, a software-defined approach to enhance security. EGEON consists of two main components: a control plane with a centralized controller for policy application and function orchestration, and a data plane with serverless functions for data flow integration, including encryption operations. This structure allows for dynamic and flexible security management, protecting digital data against external and internal threats.