MS-Stream: A multiple-source adaptive streaming solution enhancing consumer's perceived quality

Abstract

Adaptive bitrate streaming protocols, such as DASH, have seen extensive interests for their adaptation capabilities to increase consumers' Quality of Experience (QoE) over the Internet, and have become de-facto standards in web video delivery. Compared with traditional single-server approaches, multipleserver streaming offers the opportunity to exploit expanded bandwidth, link diversity, and reliability. In this paper, we expose our solution for multiple-server support to dynamic adaptive streaming applications: Multiple-Source Streaming (MS-Stream). Thanks to its codec agnosticism and DASH-compliance our contribution is a pragmatic and evolving solution for QoE enhancement that can be applied to many streaming architectures (CDNs, Clouds) and is particularly suited for distributed environments such as P2P or Set-Top-Box overlays. In addition, splitting content into multiple independent sub-streams provides the opportunity to achieve easy-to-design bitrate adaptation and server-switching mechanisms. We empirically validate our approach using an extensive collection of network profiles provided by the DASH Industry Forum. Our solution is compared with the full potential of DASH with several servers over several QoE criteria. Results show the QoE gain of using MS-Stream against DASH; an online demonstration is made available.

Adding a New Dimension to HTTP Adaptive Streaming Through Multiple-Source Capabilities

“…Although most of the proposed MDC approaches [5] rely on two descriptions only, our solution can generate any number, enabling the use of an important number of servers, providing better bitrate scalability and resilience to outages and delay variations [5]. The work [11][12] also relies on the use of descriptions. However, descriptions in [ [11][12] are created prior the content consumption phase that limits network heterogeneity adaptability to the number of available descriptions, leading to many quality fluctuations.…”

“…The work [11][12] also relies on the use of descriptions. However, descriptions in [ [11][12] are created prior the content consumption phase that limits network heterogeneity adaptability to the number of available descriptions, leading to many quality fluctuations. Moreover, description placement in a distributed system is an obvious issue of [11][12], requiring expanded storage capacity.…”

“…However, descriptions in [ [11][12] are created prior the content consumption phase that limits network heterogeneity adaptability to the number of available descriptions, leading to many quality fluctuations. Moreover, description placement in a distributed system is an obvious issue of [11][12], requiring expanded storage capacity. In our work, descriptions are generated for each video segment, during consumption phase, and upon client specification.…”

Clustering‐based quality selection heuristics for HTTP adaptive streaming over cache networks

“…These works generally focus on the simultaneous download of content from multiple servers, with recent examples including MS-Stream and MUSLIN. 9,10 In these works, streams from different servers are combined in order to improve the overall throughput and, overall, increase the user's QoE. In our work, however, the client is assumed to be unaware of the existence of multiple sources: transparent intermediate caches are used to intercept requests and deliver the requested content if possible.…”

QOE enhancement through cost-effective adaptation decision process for multiple-server streaming over HTTP

“…In contrast with [7] and [8], the segments retrieved in our approach can be independent from each other and are aggregatable, thus providing uninterrupted streaming even when segments are lost. The work in [9] [10] relies on the same principles presented in our work (i.e., independent and aggregatable sub-segments). However, due to the fact that subsegments in [9] [10] are created prior the content consumption phase, flexibility and adaptability to network paths heterogeneity are drastically limited to the number of available sub-segments, leading to many quality fluctuations throughout streaming sessions.…”