To DASH, or Not to DASH? Optimal Video Bitrate Selection and Edge Network Caching in MEC-Empowered Slice-Enabled Networks
Dionysis Xenakis
Abstract
An increasing body of works highlight that Dynamic adaptive streaming over HTTP (DASH) can negatively impact the user Quality of Experience (QoE) in cache-enabled networks. Besides, network-agnostic DASH cannot fully harness the potential of slice-enabled 5G and Beyond networks, which can promise minimum rate performance for the i) MEC-empowered cellular base station (BS) to user link, and ii) content delivery network (CDN) to MEC-empowered BS link. In this paper, we show that network-agnostic DASH is incompatible with MEC-empowered edge network caching and network slicing in next generation mobile data networks. Accordingly, we propose joint video bitrate selection and edge network caching when specific rate and cache size guarantees can be provided along the end-to-end content delivery path. By formulating the aforementioned problem as a dynamic program (DP), we present an exhaustive (brute-force) search algorithm to optimally solve it and derive an exact algorithm of polynomial time to avoid unnecessary recalculations of previously visited solution branches. Through extensive experimental results we assess the performance of the proposed algorithm and compare it against other solutions, aiming to draw valuable insights for algorithmic design tailored to MEC-empowered slice-enabled mobile data networks.