Optimal Delivery of Rate-Adaptive Streams in Underprovisioned Networks

Abstract

Abstract-The growth of Internet video traffic imposes a severe capacity problem in today's Content Delivery Network (CDN). Rate-adaptive streaming technologies, such as the Dynamic Adaptive Streaming over HTTP (DASH) standard, reinforces this problem in the core CDN infrastructure since delivering one video means delivering multiple representations for an aggregated bit-rate that is commonly over 10 Mbps. In this paper, we explore better trade-offs between CDN infrastructure cost and Quality of Experience (QoE) of the end-users for live broadcast video streaming applications. We consider in particular underprovisioned CDN networks, our goal being to maximize the QoE for the population of heterogeneous end-users despite the lack of resources in the intermediate CDN equipments. We show that previous theoretical models based on elastic bit-rates do not fit for this context. We propose a user-centric discretized streaming model where the satisfaction of end-users is related to the context and where a stream has to be either delivered in its entirety, or not delivered at all. We first formulate an Integer Linear Program (ILP) that achieves the optimal delivery through a multi-tree delivery overlay. The evaluation of the ILP shows the benefits of this model. We then design a practical system by revisiting the three main algorithms implemented in CDN: userto-server assignment, content placement and content delivery. At last, we use a realistic trace-driven large-scale simulator to study the performances of our system. In particular, we show that the population of users is reasonably well served (three quarters of the population do not experience degradation) even when the CDN infrastructure experiences a severe underprovisioning (less than half of the required infrastructure).

Instantaneous throughput prediction in cellular networks: Which information is needed?

“…Some researchers have also studied the case where the core network [18] or the CDN [19] are under-provisioned, in which case the bottleneck is not in the last-mile. In this case, predicting the data rate on the wireless link is not enough to predict the end-to-end rate but such a prediction can still be exploited by combining it with information related to the status of the core network and the CDN.…”

Joint Optimization for the Delivery of Multiple Video Channels in Telco-CDNs

“…The channel importance is generally related to the popularity of the content in the area. We assume all channels have the same bitrate, and do not consider rate-adaptive video streaming systems like [26] where video bitrates are heterogeneous. But the stream of each channel can be extended to "a bundle of streams", which contains all the "representations" of the same video channel.…”

Optimal tree packing for discretized live rate-adaptive streaming in CDN

“…The real value of the utility scores can be determined according to different application environments. For example, in [22] edge servers exclusively serving mobile users can set high utility scores for low representations to avoid receiving the useless High-Definition (HD) representations. Whereas edge servers serving Table 1 Notations Utility score of edge server u to d ij high-bandwidth users give high priority for HD representations to guarantee user Quality of Experience (QoE).…”

“…The utility shows the preferences of edge servers on representations, thus the definition of utility could drive the delivery process. For example, in [22], by setting the utilities into proper values, edge servers can be specified into receiving some…”