Conference Paper

On the effectiveness of video prefetching relying on recommender systems for mobile devices

If you want to read the PDF, try requesting it from the authors.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The intuitive idea of prefetching recent videos from subscribed YouTube channels is not sufficient, as we will show in the data analysis section of this paper. Also selecting videos for prefetching recommended by YouTube [11] and the video's like count [12] have shown a poor performance. Hence, we conclude that personalized prefetching of videos is still a challenging task since native features, such as the subscription status and the global popularity, are usually rather ineffective for prefetching algorithms. ...
... This results in 27 user devices, in the following denoted as participants or users. Thereby, the number of our participants and duration surpasses the related works with 10-15 users observed over 10 days -8 weeks [11], [12], [17]. Statistics on our participants are shown in Table I showing that users are quite diverse w.r.t. the number of days they participated in the study, the number of the subscribed channels (#Subscriptions), and the number of channels they requested videos from (#Channels watched). ...
... The results significantly differ, i.e., the median BHR was almost 0 as pseudo subscriptions are the dominant source of videos for many participants. Summarizing, vFetch outperforms existing prefetching mechanisms with a BHR between 0.3% and 14% compared with ≤ 0.03% [11], [16], [22] when applied to YouTube as shown in [11]. ...
... In view of (19) and (20), the steady-state probability vector π can be obtained as a solution of the system πT = π (21) where the transition matrix T of size (L + 1) × (L + 1) contains the transition probabilities π i,j , i.e., ...
Article
Full-text available
The emerging 5G mobile networks are essential enablers for mobile virtual reality (VR) video streaming applications assuring high quality of experience (QoE) at the end-user. In addition, mobile edge computing brings computational resources closer to the user equipment (UE), which allows offloading computationally intensive processing. In this paper, we consider a network architecture for mobile VR video streaming applications consisting of a server that holds the VR video content, a mobile edge virtualization with prefetching (MVP) unit that handles the VR video packets, and a head-mounted display along with a buffer, which together serve as the UE. Several modulation and coding schemes with different rates are provided by the MVP unit to adaptively cope with the varying wireless link conditions to the UE and the state of the UE buffer. The UE buffer caches VR video packets as needed to compensate for the adaptive rates. A performance analysis is conducted in terms of blocking probability, throughput, queueing delay, and average packet error rate. To capture the effect of fading severity, the analytical expressions for these performance metrics are derived for Nakagami-m fading on the wireless link from the MVP unit to the UE. Numerical results show that the proposed system meets the network requirements needed to assure the QoE levels of different mobile VR video streaming applications.
... While it is easy to predict that a user will likely watch the next episode of a TV series on Netflix, for example, this does not hold in the case of music consumption. In fact, several studies have shown poor performance of approaches based on the user's subscription status and global item popularity [Koch et al. 2017b;Wilk et al. 2015Wilk et al. , 2016. ...
Preprint
Full-text available
The music domain is among the most important ones for adopting recommender systems technology. In contrast to most other recommendation domains, which predominantly rely on collaborative filtering (CF) techniques, music recommenders have traditionally embraced content-based (CB) approaches. In the past years, music recommendation models that leverage collaborative and content data -- which we refer to as content-driven models -- have been replacing pure CF or CB models. In this survey, we review 47 articles on content-driven music recommendation. Based on a thorough literature analysis, we first propose an onion model comprising five layers, each of which corresponds to a category of music content we identified: signal, embedded metadata, expert-generated content, user-generated content, and derivative content. We provide a detailed characterization of each category along several dimensions. Second, we identify six overarching challenges, according to which we organize our main discussion: increasing recommendation diversity and novelty, providing transparency and explanations, accomplishing context-awareness, recommending sequences of music, improving scalability and efficiency, and alleviating cold start. Each article addressing one or more of these challenges is categorized according to the content layers of our onion model, the article's goal(s), and main methodological choices. Furthermore, articles are discussed in temporal order to shed light on the evolution of content-driven music recommendation strategies. Finally, we provide our personal selection of the persisting grand challenges, which are still waiting to be solved in future research endeavors.
... • Pre-fetching at the edge: Many solutions are also proposed regarding pre-fetching of video streams on client or networkside. Approaches utilized on client-side includes downloading multiple segments in parallel through multi-path TCP [12], and a recommender that predicts videos likely to be watched by users [13]. Note that client-side video pre-fetching strategies do not wisely consider end-to-end path between Radio Access Networks (RAN) and the cloud, hence experience sub-optimal TCP performance. ...
Article
Today's Internet traffic is dominated by video streaming applications transmitted through wireless/cellular interfaces of mobile devices. Although ultra-high-definition videos are now easily transmitted through mobile devices, video quality level that users perceive is generally lower than expected due to distance-based high latency between sources and end-users. Mobile edge computing (MEC) paradigm is expected to address this issue and provide users with higher perceived Quality of Experience (QoE) for latency-critical applications, deploying MEC servers at edges. However, due to capacity concerns on MEC servers, a more comprehensive approach is needed to meet user's expectations applying all possible operations over the resources such as caching, pre-fetching and task offloading policies depending on the data repetition or memory/CPU utilization. To address these issues, this paper proposes a novel collaborative QoE-based mobility-aware video streaming scheme deployed at a MEC server. Throughout the paper, we demonstrate how the proposed scheme can be implemented so as to preserve the desired QoE level per user during entire video sessions. Performance of the proposed scheme has been investigated by extensive simulations. In comparison to existing schemes; the results illustrate that high efficiency is achieved through collaboration among MEC servers, utilizing explicit window size adaptation, collaborative pre-fetching and handover among the edges.
... In [19] and [20], the authors proposed to prefetch different segments of either one video or multiple videos in parallel through multipath TCP. In [21], it is proposed to pre-download YouTube videos to UEs based on a recommender system, which predicts which video(s) are likely to be watched by users. In [22], the authors proposed to perform adaptive prefetching at the UE, where different prefetching strategies are calculated by considering fluctuating wireless channel conditions, memory constraints and application latency. ...
Article
Internet video streaming applications have been demanding more bandwidth and higher video quality, especially with the advent of Virtual Reality (VR) and Augmented Reality (AR) applications. While adaptive streaming protocols like MPEG-DASH (Dynamic Adaptive Streaming over HTTP) allows video quality to be flexibly adapted, e.g., degraded when mobile network condition deteriorates, this is not an option if the application itself requires guaranteed 4K quality at all time. On the other hand, conventional end-to-end TCP has been struggling in supporting 4K video delivery across long-distance Internet paths containing both fixed and mobile network segments with heterogeneous characteristics. In this paper, we present a novel and practically-feasible system architecture named MVP (Mobile edge Virtualization with adaptive Prefetching), which enables content providers to embed their content intelligence as a virtual network function (VNF) into the mobile network operator's (MNO) infrastructure's edge. Based on this architecture, we present a context-aware adaptive video prefetching scheme in order to achieve QoE-assured 4K video on demand (VoD) delivery across the global Internet. Through experiments based on a real LTE-A network infrastructure, we demonstrate that our proposed scheme is able to achieve QoE-assured 4K VoD streaming, especially when the video source is located remotely in the public Internet, in which case none of the state-of-the-art solutions is able to support such an objective at global Internet scale.
Article
The ever-increasing demand for videos on mobile devices poses a significant challenge to existing cellular network infrastructures. To cope with the challenge, we propose a user centric video transmission mechanism based on device-to-device (D2D) communications that allows mobile users to cache and share videos between each other, in a cooperative manner. The proposed solution jointly considers users' similarity in accessing videos, users' sharing willingness, users' location distribution, and users' quality of experience (QoE) requirements, in order to achieve a QoE-guaranteed video streaming service in a cellular network. Specifically, a service set consisting of several service providers and mobile users, is dynamically configured to provide timely service according to the probability of successful service. Numerical results show that when the number of providers and demanded videos is 40 and 2 respectively, the improved users experience rate in the proposed solution is approximately 85%, and the data offload rate on base station(s) is about 78%.
Conference Paper
Full-text available
The access to Online Social Networks (OSN) and to media shared over these platforms account for around 20% of today's mobile Internet traffic. For mobile device users, the access to media content and specifically videos is still challenging and costly. Mobile contracts usually have a data cap and connection qualities can vary greatly, depending on the cellular network coverage. Prefetching mechanisms that fetch content items beforehand, in times when the mobile device is connected to a WiFi network, have a high potential to address these problems. Yet, such a mechanism can only be effective if relevant content can be predicted with a high accuracy. Therefore, in this paper, an analysis of content properties and their potential for prediction are presented. An initial user study with 14 Facebook users running an app on their mobile device was conducted. The results show that video consumption is very diverse across the users. This work discusses the evaluation setup, the data analysis, and their potential to define an effective prefetching algorithm.
Conference Paper
Full-text available
We discuss the video recommendation system in use at YouTube, the world's most popular online video community. The system recommends personalized sets of videos to users based on their activity on the site. We discuss some of the unique challenges that the system faces and how we address them. In addition, we provide details on the experimentation and evaluation framework used to test and tune new algorithms. We also present some of the findings from these experiments.
Article
Full-text available
Web caching is an important technique to scale the Internet. One important performance factor of Web caches is the replacement strategy. Due to specific characteristics of the World Wide Web, there exist a huge number of proposals for cache replacement. This article proposes a classification for these proposals that subsumes prior classifications. Using this classification, different proposals and their advantages and disadvantages are described. Furthermore, the article discusses the importance of cache replacement strategies in modern proxy caches and outlines potential future research topics.
Conference Paper
Huge quantities of videos are shared via Online Social Networks (OSN) like Facebook and are watched on mobile devices. Internet connections via cellular networks (UMTS / LTE) require the scarce resources radio bandwidth and battery power. Prefetching of videos in areas of WLAN availability has the potential to reduce the power consumption in comparison to data transmission via cellular networks and prefetching can help to avoid users running into traffic caps of their network providers. Furthermore, startup delays can be reduced. Social networks offer contextual information such as likes and comments as well as social graph information which can potentially be used to predict which content will be consumed in the near future. In this paper, we elaborate possibilities to predict content consumption based on the number of likes, comments and the social graph distance. Our detailed analysis of the media access patterns of more than 700 users in Facebook shows that the media consumption does not solely depend on the number of likes or comments. Users tend to watch videos that are uploaded by close friends and family members. Furthermore, the time a video preview stays in the browser-viewport before being clicked (pre-click delay) can be exploited to decrease startup delays.
Article
The distribution of videos over the Internet is drastically transforming how media is consumed and monetized. Content providers, such as media outlets and video subscription services, would like to ensure that their videos do not fail, start up quickly, and play without interruptions. In return for their investment in video stream quality, content providers expect less viewer abandonment, more viewer engagement, and a greater fraction of repeat viewers, resulting in greater revenues. The key question for a content provider or a content delivery network (CDN) is whether and to what extent changes in video quality can cause changes in viewer behavior. Our work is the first to establish a causal relationship between video quality and viewer behavior, taking a step beyond purely correlational studies. To establish causality, we use Quasi-Experimental Designs, a novel technique adapted from the medical and social sciences. We study the impact of video stream quality on viewer behavior in a scientific data-driven manner by using extensive traces from Akamai's streaming network that include 23 million views from 6.7 million unique viewers. We show that viewers start to abandon a video if it takes more than 2 s to start up, with each incremental delay of 1 s resulting in a 5.8% increase in the abandonment rate. Furthermore, we show that a moderate amount of interruptions can decrease the average play time of a viewer by a significant amount. A viewer who experiences a rebuffer delay equal to 1% of the video duration plays 5% less of the video in comparison to a similar viewer who experienced no rebuffering. Finally, we show that a viewer who experienced failure is 2.32% less likely to revisit the same site within a week than a similar viewer who did not experience a failure.
Article
With the recent advent of 4G LTE networks, there has been increasing interest to better understand the performance and power characteristics, compared with 3G/WiFi networks. In this paper, we take one of the first steps in this direction. Using a publicly deployed tool we designed for Android called 4GTest attracting more than 3000 users within 2 months and extensive local experiments, we study the network performance of LTE networks and compare with other types of mobile networks. We observe LTE generally has significantly higher downlink and uplink throughput than 3G and even WiFi, with a median value of 13Mbps and 6Mbps, respectively. We develop the first empirically derived comprehensive power model of a commercial LTE network with less than 6% error rate and state transitions matching the specifications. Using a comprehensive data set consisting of 5-month traces of 20 smartphone users, we carefully investigate the energy usage in 3G, LTE, and WiFi networks and evaluate the impact of configuring LTE-related parameters. Despite several new power saving improvements, we find that LTE is as much as 23 times less power efficient compared with WiFi, and even less power efficient than 3G, based on the user traces and the long high power tail is found to be a key contributor. In addition, we perform case studies of several popular applications on Android in LTE and identify that the performance bottleneck for web-based applications lies less in the network, compared to our previous study in 3G [24]. Instead, the device's processing power, despite the significant improvement compared to our analysis two years ago, becomes more of a bottleneck.
Article
Video sharing has been an increasingly popular application in online social networks (OSNs). However, its sustainable development is severely hindered by the intrinsic limit of the client/server architecture deployed in current OSN video systems, which is not only costly in terms of server bandwidth and storage but also not scalable. The peer-assisted Video-on-Demand (VOD) technique, in which participating peers assist the server in delivering video content has been proposed recently. Unfortunately, videos can only be disseminated through friends in OSNs. Therefore, current VOD works that explore clustering nodes with similar interests or close location for high performance are suboptimal, if not entirely inapplicable, in OSNs. Based on our long-term real-world measurement of over 1,000,000 users and 2,500 videos in Facebook, we propose SocialTube, a novel peer-assisted video sharing system that explores social relationship, interest similarity, and physical location between peers in OSNs. Specifically, SocialTube incorporates three algorithms: a social network (SN)-based P2P overlay construction algorithm, a SN-based chunk prefetch algorithm, and a buffer management algorithm. The trace driven based simulation results show that SocialTube can improve the quality of user experience and system scalability over current P2P VOD techniques.
Conference Paper
The recent three years have witnessed an explosion of networked video sharing, represented by YouTube, as a new killer Internet application. Their sustainable development however is severely hindered by the intrinsic limit of their client/server architecture. A shift to the peer-to-peer paradigm has been widely suggested with success already shown in live video streaming and movie-on-demand. Unfortunately, our latest measurement demonstrates that short video clips exhibit drasti- cally different statistics, which would simply render these existing solutions suboptimal, if not entirely inapplicable. Our long-term measurement over five million YouTube videos, on the other hand, reveal interesting social networks with strong clustering among the videos, thus opening new opportunities to explore. In this paper, we present NetTube, a novel peer-to- peer assisted delivering framework that explores the clustering in social networks for short video sharing. We address a series of key design issues to realize the system, including a bi-layer overlay, an efficient indexing scheme and a pre-fetching strategy leveraging social networks. We evaluate NetTube through simulations and prototype experiments, which show that it greatly reduces the server workload, improves the playback quality and scales well.
Search Evaluation at Google
  • S Huffman
S. Huffman, "Search Evaluation at Google," Tech. Rep., Sep. 2008. [Online]. Available: http://googleblog.blogspot.de/2008/09/searchevaluation-at-google.html, Checked at: 20141113
A close examination of performance and power characteristics of 4g lte networks
  • J Huang
  • F Qian
  • A Gerber
  • Z M Mao
  • S Sen
  • O Spatscheck
J. Huang, F. Qian, A. Gerber, Z. M. Mao, S. Sen, and O. Spatscheck, "A close examination of performance and power characteristics of 4g lte networks," in ACM International Conference on Mobile Systems, Applications, and Services, 2012.