GNN-GM: A Proactive Caching Scheme for Named Data Networking

Abstract

As people spend more time watching movies and sharing videos online, it is crucial to provide users with a satisfactory quality of experience (QoE). With the help of the in-network caching feature in named data networking (NDN), our paper aims to improve user experience through caching. We propose a graph neural network-gain maximization (GNN-GM) cache placement algorithm. First, we use a GNN model to predict users’ ratings of unviewed videos. Second, we consider the total predicted rating of a video as the gain of caching the video. Third, we propose a cache placement algorithm to maximize the caching gains and proactively cache videos. We also design a caching replacement strategy based on the gain of caching the video. We utilize a real-world dataset to evaluate our caching strategy. Compared to state-of-the-art caching approaches, experimental results show that our caching policy improves cache hit rate by 25%, reduces latency by 5%, and reduces server load by 7%.