Integrated prefetching and caching for adaptive video streaming over HTTP

Liang, Ke; Hao, Jialei; Zimmermann, Roger; Yau, David K. Y.

doi:10.1145/2713168.2713181

Cited by 26 publications

(17 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [6], Liang et al combine both caching and pre-fetching to improve the performance in terms of byte-hit ratio and video bit rates. The architecture of this solution is composed of three modules: a cache manager, a pre-fetch manager and a request pool.…”

Section: State Of the Artmentioning

confidence: 99%

“…The purpose of this is to make the network less vulnerable to congestion by making the same content available in places other than the server, so there are fewer requests sent directly to the server, and users can get better performance by streaming the content from the closest available cache. There have also been some attempts to study the impact of pre-fetching on user QoE [6], [7]; however, the cache often needs a considerably large size in order to yield an acceptable cache hit percentage.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Just-in-time proactive caching for DASH video streaming

Coutinho

Chiariotti

Zucchetto

et al. 2018

2018 17th Annual Mediterranean Ad Hoc Networking Workshop (Med-Hoc-Net)

View full text Add to dashboard Cite

Since video traffic has become the major source of generated mobile traffic worldwide, there has been a strong interest in new techniques to improve the Quality of Experience (QoE) of users generating this traffic. Several efforts have been made to propose adaptive streaming algorithms on the client side to be able to adjust the quality of the downloaded video segments to the current network conditions, taking advantage of the Dynamic Adaptive Streaming over HTTP (DASH) standard. This study uses this knowledge to propose a pre-fetching proxy cache, to be placed at the network's edge, which will predict the quality that the client will request for the following segment. The proxy predicts the future network conditions and models the system as a Markov Decision Process (MDP), in order to find the optimal decision for the proxy, given the current network conditions. This Just-in-Time caching technique pre-fetches the segment just before the client requests it, aiming to decrease the total time spent by the client downloading the segments, and indirectly increasing the user's QoE, as the DASH client will perceive better network conditions. The study concludes that the predictive technique improves QoE by increasing the video quality and decreasing the number of stalling events, in comparison to solutions that pre-fetch the previously requested quality, and solutions which do not use any pre-fetching technique.

show abstract

Section: State Of the Artmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Just-in-time proactive caching for DASH video streaming

Coutinho

Chiariotti

Zucchetto

et al. 2018

2018 17th Annual Mediterranean Ad Hoc Networking Workshop (Med-Hoc-Net)

View full text Add to dashboard Cite

show abstract

“…Caching schemes which use pre-fetching have been shown to be beneficial for TV-on-demand and VoD services [9,13,15,17,22]. To the best of our knowledge, none of the existing works have attempted to model the request arrival process for VoD services with recommendation systems, and instead, use trace data to evaluate the performance of the proposed policies.…”

Section: Pre-fetching Based Caching Schemesmentioning

confidence: 99%

“…The goal of reducing start-up delay motivates caching policies that are aggressive in adapting the content stored on the local servers in order to minimize the probability of delayed start-up. One possible solution is to pre-fetch (fetch before videos are requested) and cache videos that are likely to be requested in the near future [9,13,15,17,22]. Since pre-fetching leads to an increase in the bandwidth consumption of the CDN, there is a trade-off between bandwidth usage of the network and the quality of service provided to the users.…”

Section: Introductionmentioning

confidence: 99%

Modeling Request Patterns in VoD Services with Recommendation Systems

Gupta

Moharir

2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Video on Demand (VoD) services like Netflix and YouTube account for ever increasing fractions of Internet traffic. It is estimated that this fraction will cross 80% in the next three years. Most popular VoD services have recommendation engines which recommend videos to users based on their viewing history, thus introducing time-correlation in user requests. Understanding and modeling this time-correlation in user requests is critical for network traffic engineering. The primary goal of this work is to use empirically observed properties of user requests to model the effect of recommendation engines on request patterns in VoD services. We propose a Markovian request model to capture the timecorrelation in user requests and show that our model is consistent with the observations of existing empirical studies. Most large-scale VoD services deliver content to users via a distributed network of servers as serving users requests via geographically co-located servers reduces latency and network bandwidth consumption. The content replication policy, i.e., determining which contents to cache on the servers is a key resource allocation problem for VoD services. Recent studies show that low start-up delay is a key Quality of Service (QoS) requirement of users of VoD services. This motivates the need to prefetch (fetch before contents are requested) and cache content likely to be requested in the near future. Since pre-fetching leads to an increase in the network bandwidth usage, we use our Markovian model to explore the trade-offs and feasibility of implementing recommendation based prefetching.

show abstract

“…One possible solution is to pre-fetch (fetch before videos are requested) and cache videos that are likely to be requested in the near future [9,13,15,17,22]. Since pre-fetching leads to an increase in the bandwidth consumption of the CDN, there is a trade-off between bandwidth usage of the network and the quality of service provided to the users.…”

Section: Introductionmentioning

confidence: 99%

Request patterns and caching for VoD services with recommendation systems

Gupta¹,

Moharir²

2017

2017 9th International Conference on Communication Systems and Networks (COMSNETS)

View full text Add to dashboard Cite

Abstract. Video on Demand (VoD) services like Netflix and YouTube account for ever increasing fractions of Internet traffic. It is estimated that this fraction will cross 80% in the next three years. Most popular VoD services have recommendation engines which recommend videos to users based on their viewing history, thus introducing time-correlation in user requests. Understanding and modeling this time-correlation in user requests is critical for network traffic engineering. The primary goal of this work is to use empirically observed properties of user requests to model the effect of recommendation engines on request patterns in VoD services. We propose a Markovian request model to capture the timecorrelation in user requests and show that our model is consistent with the observations of existing empirical studies. Most large-scale VoD services deliver content to users via a distributed network of servers as serving users requests via geographically co-located servers reduces latency and network bandwidth consumption. The content replication policy, i.e., determining which contents to cache on the servers is a key resource allocation problem for VoD services. Recent studies show that low start-up delay is a key Quality of Service (QoS) requirement of users of VoD services. This motivates the need to prefetch (fetch before contents are requested) and cache content likely to be requested in the near future. Since pre-fetching leads to an increase in the network bandwidth usage, we use our Markovian model to explore the trade-offs and feasibility of implementing recommendation based prefetching.

show abstract

Integrated prefetching and caching for adaptive video streaming over HTTP

Cited by 26 publications

References 27 publications

Just-in-time proactive caching for DASH video streaming

Just-in-time proactive caching for DASH video streaming

Modeling Request Patterns in VoD Services with Recommendation Systems

Request patterns and caching for VoD services with recommendation systems

Contact Info

Product

Resources

About