Implementation of edge-AI and cloud-gaming benchmarks for cloud-edge systems

• Main Author: Lim, Yan Kai
• Other Authors: Dmitrii Ustiugov
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; Distributed Systems
• Online Access: https://hdl.handle.net/10356/175155
• Institution: Nanyang Technological University

This report seeks to document the development process and the design decisions behind the Edge-AI and Cloud Gaming benchmarks. The benchmarks are under an open-sourced initiative, the vHive ecosystem, which is a research platform to analyze serverless systems. The report begins with an overview of cloud and edge computing, followed by the benefits of serverless applications. Subsequently the vHive ecosystem is introduced before delving into the architecture design of both the benchmarks. The current benchmarks developed is meant to be a solid foundation for vHive in demonstrating implementations of high performing systems using the serverless paradigm. Therefore this report also aims to be a guide to maintaining and improving the existing benchmarks so that it simulates industrial standard systems.