A Generic Approach to Video Buffer Modeling Using Discrete-Time Analysis

Abstract

The large share of traffic in the Internet generated by video streaming services puts high loads on access and aggregation networks, resulting in high costs for the content delivery infrastructure. To reduce the bandwidth consumed while maintaining a high playback quality, video players use policies that control and limit the buffer level by using thresholds for pausing and continuing the video download. This allows shaping the bandwidth consumed by video streams and limiting the traffic wasted in case of playback abortion. Especially in mobile scenarios, where the throughput can be highly variant, the buffer policy can have a high impact on the probability of interruptions during video playback. To find the optimal setting for the buffer policy in each network condition, the relationship between the parameters of the buffer policy, the network throughput dynamics, and the corresponding video playback behavior needs to be understood. To this end, we model the video buffer as GI/GI/1 queue with pq -policy using discrete-time analysis. By studying the stochastic properties of the buffer-level distribution, we are able to accurately evaluate the impact of network and video bitrate dynamics on the video playback quality based on the buffer policy. We find a fundamental relationship between the bandwidth variation and the expected interarrival time of segments, meaning that overproportionately more bandwidth is necessary to prevent stalling events for high bandwidth variation. The proposed model further allows to optimize the trade-off between the traffic wasted in case of video abortion and video streaming quality experienced by the user.

Quality of Experience for Streaming Services: Measurements, Challenges and Insights

“…It is proved to adapt the quality of the requested video, based on the current bandwidth and devices qualification, but it is affected by many factors based on [161], [162], initial delay, stalling and level variation (frame rate, bit-rate and resolution), besides other factors like video length and the number of motions in the video. Consequently, to derive an effective the trade-off between the network variations and dynamic videos streaming behavior, they [163] introduce a queue-based model to analyze the video buffer (GI/GI/1 queue) with pq-policy (pausing or continuing the video download) using discrete-time analysis. Suggesting to adjust the buffering thresholds according to the bandwidth fluctuations to reduce the stalling vents.…”

Computing QoE-Relevant Adaptive Video Streaming Metrics Using Discrete-Time Analysis

“…Burger et al [13] models the video buffer as a GI/GI/1 queue with pq-policy using discrete time-analysis. Thereby, the video portion buffered at the client is considered as the amount of unfinished work in the system.…”

“…Instead, such comparisons are done for specific use-cases which are considered to be relevant. Recently a couple of queueingbased models [13], [14] have been developed. These models are based on certain assumptions regarding the adaptation strategy and other relevant parameters, but allow to easily compute QoE metrics like the stalling probability for a large set of different network scenarios and parameter settings.…”

“…To close this gap, we enhance a discrete-time model from literature [13] by including explicit quality switching based on the state of the video buffer. Firstly, we present a formal discrete-time model for adaptive streaming systems.…”

Evaluation of the Benefits of Variable Segment Durations for Adaptive Streaming

“…A model for investigating the stochastic properties of the buffer level distribution is introduced in [14]. The model is based on a GI/GI/1 queue with pq-policy and allows the computation of relevant performance metrics like the stalling probability, the stalling duration or the buffer utilization based on distributions for the segment duration, the segment size and the bandwidth.…”