Follow the User: A Framework for Dynamically Placing Content Using 5G-Enablers

Abstract

This article presents a framework for improved and efficient video delivery in scenarios featuring users moving at high speed (e.g., trains), leveraging on dynamic Multi-access Edge Computing (MEC) enabled 5G network capabilities. The framework is location-aware and it allows the content to efficiently follow the users, conserving load usage on network and computational resources, by placing virtualized Content Delivery Network (vCDN) nodes at edge sites. The nodes are controlled by the framework's centralized control unit, which is able to dynamically and preemptively deploy virtulized resources, as the train moves. The framework is capable of segmenting video content and placing the specific portion of content that a user is likely to consume across a set of dynamically deployed vCDN nodes, associated to the coverage section the train is currently passing. A proof of concept was implemented and evaluated, where the benefits of using this framework are assessed. Results showed that the proposed system was able to reduce the load on the core network by 10.9 percent and maximize the cache hit ratio to a value of 99.8 percent.