Virtual reality has been gaining popularity in recent years fueled by the proliferation of affordable consumer-grade devices such as Oculus Rift, HTC Vive, and Samsung VR. Amongst the various VR applications, 360° video streaming is currently one of the most popular ones. However, it poses a series of challenges to the serving content distribution systems. One challenge is the significantly increased bandwidth requirement for streaming such content in real time. Recent research has shown that only streaming the content that is in the user's (field-of-view) FoV in high quality can lead to strong bandwidth savings. This can be achieved by analyzing the viewers head orientation and movement based on sensor information. Alternatively, historic information from users that watched the content in the past can be considered to prefetch 360° video data in high quality assuming the viewer will direct the FoV to these areas. This paper presents a 360° video streaming system that transitions between sensor- and content-based predictive mechanisms. We evaluate the effects of our system on the Quality of Experience (QoE) of such a VR streaming system and show that the perceived quality can be increased between 50% and 80% compared to systems that only apply either one of the two approaches.

Music belongs to one of the most popular content categories overall, and it is nowadays mainly consumed using online streaming services. With YouTube being the largest source of traffic in most networks about half of all YouTube requests address music videos. To cope with the continuously growing demand for content and thus increasing network traffic, YouTube operates its own CDN, a globally distributed network of caches. This allows serving content from locations close to the users, which circumvents potential network bottlenecks and increases the user-perceived QoE due to reduced latency. Recently, proactive caching and prefetching has shown superior performance results compared with traditional reactive caching schemes such as LRU and LFU. Due to the substantial footprint of music videos on today's Internet, we propose a novel proactive caching strategy specifically for music videos. This strategy incorporates two key observations: i) Music genre and mood popularity varies over the course of the day and ii) A video's past views are predictive for its future popularity development. For the classification task, we use a Convolutional Neural Network while investigating several predictive models for the popularity estimation. The proposed caching system can increase the cache hit rate up to 4.5% which is substantial for caching systems.

Content delivery networks (CDNs) carry more than half of the video content in today's Internet. By placing content in caches close to the users, CDNs help increasing the Quality of Experience, e.g., by decreasing the delay until a video playback starts. Existing works on CDN cache performance focus mostly on distinct caching metrics, such as hit rate, given an abstract workload model. Moreover, the nature of the geographical distribution and connection of caches is often oversimplified. In this work, we investigate the performance of cache hierarchies while taking into account the presence of a mixed content workload comprising multiple categories, e.g., news, comedy, and music. We consider the performance of existing caching strategies in terms of cache hit rate and deterioration costs in terms of write operations. Further, we contribute a design and an evaluation of a content category-aware caching strategy, which has the benefit of being sensitive to changing category-specific content popularity. We evaluate our caching strategy, denoted as ACDC (Adaptive Content-Aware Designed Cache), using multiple caching hierarchy models, different cache sizes, and a real world trace covering one week of YouTube requests observed in a large European mobile ISP network. We demonstrate that ACDC increases the cache hit rate for certain hierarchies up to 18.39% and decreases transmission latency up to 12%. Additionally, a decrease in disk write operations up to 55% is observed.

YouTube is the most popular platform for streaming of user-generated videos. Nowadays, professional YouTubers are organized in so-called multichannel networks (MCNs). These networks offer services such as brand deals, equipment, and strategic advice in exchange for a share of the YouTubers’ revenues. A dominant strategy to gain more subscribers and, hence, revenue is collaborating with other YouTubers. Yet, collaborations on YouTube have not been studied in a detailed quantitative manner. To close this gap, first, we collect a YouTube dataset covering video statistics over 3 months for 7,942 channels. Second, we design a framework for collaboration detection given a previously unknown number of persons featured in YouTube videos. We denote this framework, for the detection and analysis of collaborations in YouTube videos using a Deep Neural Network (DNN)-based approach, as CATANA. Third, we analyze about 2.4 years of video content and use CATANA to answer research questions guiding YouTubers and MCNs for efficient collaboration strategies. Thereby, we focus on (1) collaboration frequency and partner selectivity, (2) the influence of MCNs on channel collaborations, (3) collaborating channel types, and (4) the impact of collaborations on video and channel popularity. Our results show that collaborations are in many cases significantly beneficial regarding viewers and newly attracted subscribers for both collaborating channels, often showing more than 100% popularity growth compared with noncollaboration videos.

Video streaming is responsible for the largest portion of traffic in fixed and mobile networks. Yet, forecasts expect this amount to grow further. Especially for mobile devices connected to cellular networks, high QoE video streaming can be a challenge as the user data volume is metered and eventually limited. Also, the connection quality may vary severely. Prefetching videos is an approach to mitigate this issue. Here, videos that the user is likely to watch in advance are prefetched on the user's smartphone, e.g., while he is connected to WiFi. However, this approach can only be efficient if only the videos that are interesting for the respective user are prefetched. This constitutes a major estimation and prediction challenge. To this end, this paper presents three contributions: First, a user study over multiple months that draws valuable insights on the user video request behavior. Second, we propose a novel privacy-preserving prefetching framework denoted vFetch that prefetches videos based, e.g., on the user's affinity of YouTube channels. Third, a trace-based evaluation and parameter study that demonstrates vFetch's efficiency with a hit rate of ~50% for a 50 GB cache.

The preferred channel for listening to music is shifting towards the Internet and especially to mobile networks. Here, the overall traffic is predicted to grow by 45% annually till 2021. However, the resulting increase in network traffic challenges mobile operators. As a result, methods are researched to decrease costly transit traffic and the traffic load inside operator networks using in-network and client-side caching. Additionally to traditional reactive caching, recent works show that proactive caching increases cache efficiency. Thus, in this work, a mobile network using proactive caching is assumed. As music represents the most popular content category on YouTube, this work focuses on studying the potential of proactively caching content of this particular category using a YouTube trace containing over 4 million music video user sessions. The contribution of this work is threefold: First, music content-specific user behavior is derived and audio features of the content are analyzed. Second, using these audio features, genre and mood classifiers are compared in order to guide the design of new proactive caching policies. Third, a novel trace-based evaluation methodology for music-specific proactive in-network caching is proposed and used to evaluate novel proactive caching policies to serve either an aggregate of users or individual clients.

Video constitutes the majority of traffic in ISP networks. The demand for more and higher quality videos is growing rapidly while network capacities are limited. Today, OTT video is mostly delivered over CDNs. For ISP-internal video such as IPTV channels, IP multicast provides an efficient delivery method. However, this does not scale for OTT videos due to their large number and their Zipf-distributed popularity. SDN-based multicast enables ISPs to deliver content based on efficient network-level multicast. However, so far this concept has not been applied for OTT Video-on-Demand (VoD) scenarios. To this end, this paper's contribution is threefold. First, a novel network-based multicast approach for OTT VoD delivery is presented, named VoDCast. Second, trace-based network simulations are designed and conducted. Third, the achievable bandwidth reduction and practicability of VoDCast is shown based on a two-week YouTube trace of a nation-wide mobile provider. VoDCast is able to decrease traffic volume and variance, while keeping network provider costs in terms of state overhead and state change low.

Online media services are reshaping the way video content is watched. People with similar interests tend to request same content. This provides enormous potential to predict which content users are interested in. Besides, mobile devices are commonly used to watch videos which popularity is largely driven by its social success. In this paper, we design CPSys a Central Predictor System to prefetch relevant videos for each user. To fine tune our prefetching system, we rely on a large dataset collected from a large mobile carrier in Europe. The rationale of our prefetching strategy is first to form a graph and build implicit or explicit ties between similar users. On top of this graph, we propose the Most Popular and Most Recent (MPMR) policy to predict relevant videos for each user. We show that CPSys can achieve high performance with respect to the correct prediction ratio and by significantly reducing the traffic overhead. We further show that CPSys outperforms other prefetching schemes that have been presented and studied in the state of the art. At the end, we provide a proof-of-concept implementation of our prefetching system.

Mobile network operators are expected to face significant traffic increase in the upcoming years. One alternative method is to intelligently move transmissions to times of network underutilization, either on 3G/4G or by offloading to WiFi. Video content, predicted by Cisco to constitute 69% of mobile traffic, offers the greatest potential for offloading. To this end, the demonstrated app strives to relieve the mobile network in a two ways. First, long-term prefetching of promising videos based on posts from the user's Online Social Network feed is performed. The knowledge about which video is likely being requested in the near future offers the opportunity to schedule the transmission according to its probability of being watched. Second, the approach is complemented with short-term prefetching, which is used whenever a content could not be downloaded by long-term prefetching. In this case, resources are optimized so as to maximize the communication efficiency while preserving the quality of service. The demonstrated app considers the smartphone's observed cellular network history to optimize the mobile throughput. A customized video player implements both the long-term and short-term prefetching. It reduces both the load on mobile networks, decreases playback pausing events and hereby achieves a high QoE. Thus, the player addresses both the operators' and the users' needs.

Real-time entertainment constitutes the majority of traffic in today's mobile networks. The data volume is expected to increase in the near future, whereas the mobile bandwidth capacity is likely to increase significantly slower. Especially peak hour traffic often leads to overloaded mobile networks and poor user experience. This increases costs for the mobile operator, which has to adapt to the peak demand by capacity over provisioning. The new approach proposed in this paper aims to leverage the user's context and video meta-information to unleash the potential of video prefetching. Based on observed user interactions with social networks, the videos a user consumes from social neighbours can be predicted. Moreover, the user's daily routine even enables a prediction of the time when videos are consumed as well as the network capabilities available at that point. First results show that partial prefetching based on content categories provides a potential for efficiently offloading mobile networks. Additionally, the user experience can be improved as freezing playbacks of videos can be decreased. Initial results show a high potential for category-based prefeching.