Leveraging Video Viewing Patterns for Optimal Content Placement

Hwang, Kyung-Wook; Applegate, David; Archer, Aaron; Gopalakrishnan, V.; Lee, S.; Misra, Vishal; Ramakrishnan, K. K.; Swayne, Deborah F.

doi:10.1007/978-3-642-30054-7_4

Cited by 17 publications

(12 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, we assume that the video sessions run the full length of the videos. This is, however, not the case in real VoD systems as shown by recent research [11], [17], [18]. Nevertheless, this assumption does not affect the validity of our evaluation, as we are only concerned with the popularity estimation aspect of content placement.…”

Section: Parameters Setting In Simulationmentioning

confidence: 92%

“…k in Eq. (18). It should be clear then that IPD-based object popularity estimation can be carried out using Eq.…”

Section: Optimisation Of Mipd For Cache Replacementmentioning

confidence: 99%

“…Content management in web cache systems is often considered less challenging as the steadily falling storage costs have enabled distributed web servers (e.g., proxy cache servers) to store an enormous amount of web content [15], [16]. Moreover, it is known that in VoD systems, higher performance can be potentially achieved by storing partial videos (partial caching) instead of full videos [3], [17], [18]. We do not consider such issues of VoD workload here as the focus of our work is on video popularity estimation based on the request arrival process and not based on content analysis.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Estimating Video Popularity From Past Request Arrival Times in a VoD System

et al. 2020

View full text Add to dashboard Cite

Efficient provision of Video-on-Demand (VoD) services requires that popular videos are stored in a cache close to users. Video popularity (defined by requested count) prediction is, therefore, important for optimal choice of videos to be cached. The popularity of a video depends on many factors and, as a result, changes dynamically with time. Accurate video popularity estimation that can promptly respond to the variations in video popularity then becomes crucial. In this paper, we analyze a method, called Minimal Inverted Pyramid Distance (MIPD), to estimate a video popularity measure called the Inverted Pyramid Distance (IPD). MIPD requires choice of a parameter, k, representing the number of past requests from each video used to calculate its IPD. We derive, analytically, expressions to determine an optimal value for k, given the requirement on ranking a certain number of videos with specified confidence. In order to assess the prediction efficiency of MIPD, we have compared it by simulations against four other prediction methods: Least Recency Used (LRU), Least Frequency Used (LFU), Least Recently/Frequently Used (LRFU), and Exponential Weighted Moving Average (EWMA). Lacking real data, we have, based on an extensive literature review of real-life VoD system, designed a model of VoD system to provide a realistic simulation of videos with different patterns of popularity variation, using the Zipf (heavy-tailed) distribution of popularity and a non-homogeneous Poisson process for requests. From a large number of simulations, we conclude that the performance of MIPD is, in general, superior to all of the other four methods. INDEX TERMS Popularity prediction, video-on-demand, pre-placement, request statistic, Zipf distribution, non-homogeneous Poisson process. BILL MORAN (Life Member, IEEE) received the B.Sc. degree (Hons.

show abstract

Section: Parameters Setting In Simulationmentioning

confidence: 92%

“…k in Eq. (18). It should be clear then that IPD-based object popularity estimation can be carried out using Eq.…”

Section: Optimisation Of Mipd For Cache Replacementmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Estimating Video Popularity From Past Request Arrival Times in a VoD System

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Thus VoD content typically has a playback duration of several minutes. Recent studies also show that video viewing abandonment affects the video viewing time distribution [Hwang et al 2012]. This means that the actual viewing time can be less than the video length.…”

Section: Playback Duration File Size Formats and Qualitymentioning

confidence: 99%

Saving Energy in Mobile Devices for On-Demand Multimedia Streaming -- A Cross-Layer Approach

Hoque

Siekkinen

Nurminen

et al. 2014

ACM Trans. Multimedia Comput. Commun. Appl.

View full text Add to dashboard Cite

This paper proposes a novel energy-efficient multimedia delivery system called EStreamer. First, we study the relationship between buffer size at the client, burst-shaped TCP-based multimedia traffic, and energy consumption of wireless network interfaces in smartphones. Based on the study, we design and implement EStreamer for constant bit rate and rate-adaptive streaming. EStreamer can improve battery lifetime by 3x, 1.5x and 2x while streaming over Wi-Fi, 3G and 4G respectively. • Hoque et al.over a period of time and then sent together as one burst to the client using all the available bandwidth. In this way, the WNI is kept active only for a short period of time to receive the burst, instead of keeping the interface always active. When the burst size exceeds the playback buffer, there is packet loss. Therefore, the maximum size of a burst, or equivalently the length of a burst interval, is tuned based on an acceptable range of packet loss [Korhonen and Wang 2005].Today HTTP over TCP is by far the most prevalent set of protocols used for streaming [Guo et al. 2006]. Traffic shaping saves energy with TCP-based streaming as well [Hoque et al. 2011]. TCP-based streaming differs from the case of UDP-based streaming in one major way: TCP is a reliable protocol. When the player buffer becomes full, data accumulates into the TCP receive buffer. As a result, an arbitrarily large burst, which is larger than the playback buffer, can be used without any packet loss. However, the exact impact on the energy consumption of smartphones is not well understood. In this article, we first study the interplay between the burst size and the power consumption of smartphones. Specifically, we model the energy consumption of WNIs for bursty TCP traffic. We show that the power consumption of a smartphone decreases when the receiving burst size increases and as long as the client device can accommodate the entire burst. In contrast, the power consumption rapidly increases, if the burst size is too large compared to the available buffer space at the client.As a proof of concept, we design and implement an energy-efficient multimedia delivery system called EStreamer. It determines an energy-optimal burst size so that smooth playback of the streaming applications is not distorted. EStreamer relies on standard TCP feedback from the client. We thoroughly evaluate EStreamer through measurements with real streaming services. We focus on two aspects: (a) the potential to save energy in smartphones using Wi-Fi, 3G (HSPA) and 4G (LTE) and (b) the impact on the radio access network (RAN) signaling load when using a cellular access network.Concerning the first aspect, we measure the energy savings in four different smartphones while streaming from popular streaming services via EStreamer, such as Internet radio, YouTube, and Dailymotion. The results demonstrate that the largest energy savings can be achieved with Wi-Fi (65%), followed by 4G (50-60%), after which comes 3G (38%). It is also shown that a client can achieve similar range of energy sav...

show abstract

“…Storing the most popular files results in more cache hits, Mathematical and Algorithmic Sciences Lab, France Research Center, Huawei Technologies Co. Ltd. which decreases the impact on the traffic on the core network. Nevertheless, not all the parts of a file are equally popular [3]. Hence, a natural generalization of "store the most popular files" is to split the video files into chunks and "store the most popular chunks" instead.…”

Section: Introductionmentioning

confidence: 99%