QoE Performance for DASH Videos in a Smart Cache Environment

摘要

During the past decade, Internet has seen dramatic increase in video traffic. Users expect a high quality of experience with online video streaming. For video delivery, the DASH (Dynamic Adaptive Streaming over HTTP) standard is one of the common approaches for streaming used by content providers. In order to give users a higher quality of experience, in-network caching and prefetching are useful to reduce video delivery latency with DASH-generated videos. In this paper, we present a Smart Cache framework that uses a cache prefetching scheme that prefetches segment bitrate based on forecasted throughput at the cache entity by using previous throughput values from clients. For our study, we have implemented our framework on the GENI testbed, and our results for single-client and two-client interactions show that Smart Cache increases the byte-hitrate and reduces the number of unused prefetches for cache. We also consider the impact on Quality of Experience (QoE) for each client during contention.