Docker and Kubernetes : enhance the securities of the live-stream ASR system deployment in the cloud

• Main Author: Lee, Joshua Jun Xiang
• Other Authors: Chng Eng Siong
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/156222
• Institution: Nanyang Technological University

The goal of this project is to employ relevant and current security solutions to increase the robustness and reliability of an existing automatic speech recognition (ASR) system deployment on the cloud. The implemented solutions will harden the ASR deployment, reducing both the attack surface and exploitable vulnerabilities on the system. The security solutions utilises Nginx Ingress Controller and the secret management tool, HashiCorp Key Vault. They are applied on the ASR system hosted on the Azure cloud platform using Azure Kubernetes Service (AKS) and on the Amazon Web Service (AWS) cloud platform using Elastic Kubernetes Service (EKS). An Infrastructure as Code (IaC) tool, Terraform, is also implemented to improve the deployment of the ASR system. The security solutions better secure the system’s endpoints and carry out appropriate encryption of secret data. This ensures the performance and availability of the ASR deployment services, as well as prevents potential leakage of sensitive information. This will be illustrated in the report through architectural diagrams, figures, and tables, detailing important elements of the solution and how they are implemented. The report will also include the experiments carried out to demonstrate the effectiveness of the solution.