Optimizing for Video Storage Networking With Recommender Systems

Abstract

Driven mainly by its adoption as a new media distribution platform for content providers and its ubiquitous availability for the end user's media production and consumption, the Internet is rapidly reshaping. In particular, the stakeholders in the content distribution market are considering exploiting content delivery networks (CDNs) Everywhere solution, telcos and multiple systems operators (MSOs) including AT&T, Verizon, and Comcast can now provide their customers with a whole new experience as they can now consume the content (hosted by the network provider) on fixed or mobile devices other than the TV set alone. As a consequence, these network providers can reach more customers with their content assets. Success with mobile TV offers also may be driven by the strong growth of video-enabled smartphones: 275 million video-enabled smartphones shipped in 2010 [14].At the same time, a variety of Internet video publishers, such as Netflix*, Hulu*, BBC iPlayer*, and also IntroductionThe Internet video world is undergoing a significant transformation as a wider set of content is offered for consumption over a wider set of devices.For video services offered by network providers, the following trends can be observed. Initially, many network providers deployed Internet Protocol television (IPTV) services in a walled garden environment. Apart from being a replacement for analog television (TV) distribution, IPTV also offers interactive and personalized services, including video on demand, network-based personal video recording, and time shift Apple TV* and Google TV*, have launched new video services. They are providing their content to their end users with little involvement of the network provider.Hybrid models also exist where a telco may offer a combination of digital TV and Internet video services, as described in [6]. In such a scenario, digital TV network providers will need to expand their offer by also offering access to Internet video. They have opportunities in that they can offer better quality of experience (QoE), multiscreen differentiators, have a customer base (and billing relationship), and can position themselves as the content aggregator of choice (simplicity to end user).The network provider (telco or MSO) can play a key role in both the content delivery and content discovery. As far as delivery is concerned, it is well known that content distribution networks (CDNs) can reduce the load on the Internet and within the network provider's network [7,15,16]. With respect to discovery, customers need help in finding content that interests them, especially as the amount of available content increases. Therefore, recommender systems are often integrated with video offerings to help the end user [1,3,9,13]. The use of recommenders for the purpose of caching has also been discussed in [10]. Since the network provider has a comprehensive view of the end user's content consumption, he is (potentially) well placed to recommend content. This paper combines both key roles of content discovery and content delivery...

Recomendação de Conteúdo e Desempenho de Sistemas de Cache

“…A literatura sobre o impacto de sistemas de recomendação no desempenho de caches aindaé escassa, mas está rapidamente crescendo [Verhoeyen et al, 2012, Krishnappa et al, 2015, Chatzieleftheriou et al, 2017, Sermpezis et al, 2017.…”

“…Os primeiros a considerarem a possibilidade de configurar os sistemas de caching em função dos sistemas de recomendação foram [Verhoeyen et al, 2012]. Os autores usaram-se de simulações para mostrar as potenciais vantagens de se combinar decisões de caching e de recomendação de conteúdo.…”

A QoE centric distributed caching approach for vehicular video streaming in cellular networks

“…Furthermore, Shanmugam et al in developed a system composed of helpers with low‐rate backhaul but high storage capacity caching popular video files to alleviate the bottleneck of the backhaul capacity. Considering the wireless multimedia content transmission, we always need to use the limited network resources to guarantee the best possible experience of users . A QoE‐driven channel allocation scheme for secondary users and cognitive radio network BSs was introduced in .…”

A QoE supportive distributed caching management for vehicular video streaming in cellular networks

“…Moreover, efficient and fair resource allocation for autonomous wireless multimedia users is discussed in [4]. In the multimedia content transmission, we always need to use the limited storage space to ensure the best possible experience of users [5] [6]. Therefore, it is necessary to optimally use the limited resources, reduce bandwidth on feeder links [5].…”

“…In the multimedia content transmission, we always need to use the limited storage space to ensure the best possible experience of users [5] [6]. Therefore, it is necessary to optimally use the limited resources, reduce bandwidth on feeder links [5]. [6] proposes a content cache allocation mechanism to guarantee different users' QoE requirement by transcoding each content into a set of media files with diverse playback rates.…”