Master of Science
Date of Defense
Over the past few years, Augmented Reality (AR) and Virtual Reality (VR) have emerged as highly popular technologies that demand rapid and efficient processing of data with low latency and high bandwidth, in order to enable seamless real-time interaction between users and the virtual environment. This presents challenges for network infrastructure design, which can be addressed through edge computing. However, edge computing also presents challenges, such as selecting the appropriate edge server for computing tasks in dynamic networks with rapidly changing resource availability. Named Data Networking (NDN) is a potential future Internet architecture that could provide a balanced distribution of edge services across servers, thereby preventing service disruptions. In this study, eComVes, a novel strategy that enhances ComVes, is proposed for information-centric edge applications that adopt a correction mechanism to ensure service execution on the highest resourced server. This mechanism allows users and intermediate routers to learn about the servers’ resource status directly from the server without using any explicit control messages or probing. We evaluated the performance of the eComVes against ComVes and observed an improvement in the success ratio with maintaining consistent response time, indicating an improvement in load balance across the servers.
Hoque, Sanzida, "eComVes: Enhancing ComVes using Data Piggybacking for Resource Discovery at the Network Edge" (2023). Theses. 438.