Network-aware Prefetching Method for Short-Form Video Streaming

Abstract

Recent years have witnessed the rising popularity of short-form video platforms such as TikTok. Unlike conventional videos, short-form videos are significantly shorter, and users frequently switch between content. Therefore, it is crucial to have an effective streaming method for this new video format. In this paper, we propose a resource-efficient prefetching method for short-form video streaming. Our method dynamically adjusts the amount of prefetched video data based on network throughput conditions and user viewing behaviors. Experimental results demonstrate that our method can reduce data waste by 37 ∼ 52% compared to existing methods.