Michael Zink

Characteristics of YouTube network traffic at a campus network - Measurements, models, and implications

User-Generated Content has become very popular since new web services such as YouTube allow for the distribution of user-produced media content. YouTube-like services are different from existing traditional VoD services in that the service provider has only limited control over the creation of new content. We analyze how content distribution in YouTube is realized and then conduct a measurement study of YouTube traffic in a large university campus network. Based on these measurements, we analyzed the duration and the data rate of streaming sessions, the popularity of videos, and access patterns for video clips from the clients in the campus network. The analysis of the traffic shows that trace statistics are relatively stable over short-term periods while long-term trends can be observed. We demonstrate how synthetic traces can be generated from the measured traces and show how these synthetic traces can be used as inputs to trace-driven simulations. We also analyze the benefits of alternative distribution infrastructures to improve the performance of a YouTube-like VoD service. The results of these simulations show that P2P-based distribution and proxy caching can reduce network traffic significantly and allow for faster access to video clips.

Subjective impression of variations in layer encoded videos

Layer encoded video is an elegant way to allow adaptive transmissions in the face of varying network conditions as well as it supports heterogeneity in networks and clients. As a drawback quality degradation can occur, caused by variations in the amount of transmitted layers. Recent work on reducing these variations makes assumptions about the perceived quality of those videos. The main goal of this paper respectively its motivation is to investigate the validity of these assumptions by subjective assessment. However, the paper is also an attempt to investigate fundamental issues for the human perception of layer encoded video with time-varying quality characteristics. For this purpose, we built a test environment for the subjective assessment of layer encoded video and conducted an empirical experiment in which 66 test candidates took part. The results of this subjective assessment are presented and discussed. To a large degree we were able to validate existing (unproven) assumptions about quality degradation caused by variations in layer encoded videos, however there were also some interesting, at first sight counterintuitive findings from our experiment.

Watch global, cache local: YouTube network traffic at a campus network: measurements and implications

Web services such as YouTube which allow the distribution of user-produced media have recently become very popular. YouTube-like services are different from existing traditional VoD services because the service provider has only limited control over the creation of new content. We analyze how the content distribution in YouTube is realized and then conduct a measurement study of YouTube traffic in a large university campus network. The analysis of the traffic shows that: (1) No strong correlation is observed between global and local popularity; (2) neither time scale nor user population has an impact on the local popularity distribution; (3) video clips of local interest have a high local popularity. Using our measurement data to drive trace-driven simulations, we also demonstrate the implications of alternative distribution infrastructures on the performance of a YouTube-like VoD service. The results of these simulations show that client-based local caching, P2P-based distribution, and proxy caching can reduce network traffic significantly and allow faster access to video clips.

Layer-encoded video in scalable adaptive streaming

Combining the concepts of caching and transmission control protocol (TCP)-friendly streaming of layer-encoded video bears the problem that those videos might not be cached in full quality. Therefore, we focus in this work on the scheduling of retransmissions of missing segments of a cached video in a manner that allows clients to receive the content in an improved quality. In a first step, we conducted subjective assessments of variations in layer-encoded video with the goal to validate existing quality metrics, including our own, which are based on certain assumptions. A statistical analysis of the subjective assessment validates these assumptions. We also show that the frequently used peak signal-to-noise ratio (PSNR) is not an appropriate metric for variations in layer-encoded video. With the insight from the subjective assessment we develop heuristics for retransmission scheduling and prove their applicability by conducting a series of simulations.

An end-user-responsive sensor network architecture for hazardous weather detection, prediction and response

We present an architecture for a class of systems that perform distributed, collaborative, adaptive sensing (DCAS) of the atmosphere. Since the goal of these DCAS systems is to sense the atmosphere when and where the user needs are greatest, end-users naturally play the central role in determining how system resources (sensor targeting, computation, communication) are deployed. We describe the meteorological command and control components that lie at the heart of our testbed DCAS system, and provide timing measurements of component execution times. We then present a utility-based framework that determines how multiple end-user preferences are combined with policy considerations into utility functions that are used to allocate system resources in a manner that dynamically optimizes overall system performance. We also discuss open challenges in the networking and control of such end-user-driven systems.

Closed-loop architecture for distributed collaborative adaptive sensing of the atmosphere: meteorological command and control

Distributed Collaborative Adaptive Sensing (DCAS) of the atmosphere is a new paradigm for detecting and predicting hazardous weather using a dense network of short-range, low-powered radars to sense the lowest few kilometres of the earths atmosphere. DCAS systems are collaborative in that the beams from multiple radars are actively coordinated in a sense-and-respond manner to achieve greater sensitivity, precision and resolution than possible with a single radar. DCAS systems are adaptive in that the radars and their associated computing and communications infrastructure are dynamically reconfigured in response to changing weather conditions and end-user needs. This paper describes an end-to-end DCAS architecture and evaluates the performance of the system in an operational testbed with actual weather events and end-user considerations driving the system. Our results demonstrate how the architecture is capable of real-time data processing, optimisation of radar control and sensing of the atmosphere in a manner that maximises end-user utility.

LC-RTP (loss collection RTP): reliability for video caching in the Internet

The increasing amount of audio-visual (AV) content that is offered by web sites leads to a network bandwidth and storage capacity problem. Caching is one of the techniques that can ease this problem. But even in a caching system the distribution of data (i.e. the AV content) should be bandwidth-efficient. Furthermore the delivery to the end-user must regard the restrictions implied by real-time data. This paper describes LC-RTP, an efficient and simple reliable multicast protocol that complies with RTP. Its deployment would require neither changes to the network infrastructure nor to existing end-user presentation software. It provides lossless transmission of AV content into cache servers and concurrently, lossy real-time delivery to end-users using multicast. It achieves reliability by retransmission. The traffic increase is minimal because the transmission of the AV content and any caching will take place while the end-user is served. Support for multicast in the distribution system ensures that all cache servers of a multicast group can cache an AV content while transmitting it to a consumer. Finally we present the results of long distance file transmissions in order to show that LC-RTP performs well and meets the requirements for lossless transmission of AV content.

A Distributed Algorithm for Joint Sensing and Routing in Wireless Networks with Non-Steerable Directional Antennas

In many energy-rechargeable wireless sensor networks, sensor nodes must both sense data from the environment, and cooperatively forward sensed data to data sinks. Both data sensing and data forwarding (including data transmission and reception) consume energy at sensor nodes. We present a distributed algorithm for optimal joint allocation of energy between sensing and communication at each node to maximize overall system utility (i.e., the aggregate amount of information received at the data sinks). We consider this problem in the context of wireless sensor networks with directional, non-steerable antennas. We first formulate a joint data-sensing and data-routing optimization problem with both per-node energy-expenditure constraints, and traditional flow routing/conservation constraints. We then simplify this problem by converting it to an equivalent routing problem, and present a distributed gradient-based algorithm that iteratively adjusts the per-node amount of energy allocated between sensing and communication to reach the system-wide optimum. We prove that our algorithm converges to the maximum system utility. We quantitatively demonstrate the energy balance achieved by this algorithm in a network of small, energy-constrained X-band radars, connected via point- to-point 802.11 links with non-steerable directional antennas.

Retransmission scheduling in layered video caches

In contrast to classical assumptions in video on demand (VoD) research, the main requirements for VoD in the Internet are adaptiveness, support of heterogeneity, and last but not least high scalability. Hierarchically layered video encoding is particularly well suited to deal with adaptiveness and heterogeneity support for video streaming. A distributed caching architecture is the key to a scalable VoD solution in the Internet. Thus, the combination of caching and layered video streaming is promising for an Internet VoD system, yet, requires thoughts about some new issues and challenges. In this paper, we investigate one particular of these issues: how to deal with retransmissions of missing segments for a cached layered video in order to meet user demands to watch high quality video with relatively few quality variations. We devise a suite of fairly simple retransmission scheduling algorithms and compare these against existing ones by simulative experiments.

SQUAD: a spectrum-based quality adaptation for dynamic adaptive streaming over HTTP

The application-layer based control loops of dynamic adaptive streaming over HTTP (DASH) make video bitrate selection a complex problem. In this work, we review and present new insights into the challenges of DASH rate adaptation. We identify several critical issues that contribute to the degradation of DASH performance with respect to the rate control loops of DASH and TCP. We then introduce a novel DASH quality adaptation algorithm SQUAD, which is specifically designed to ensure high quality of experience (QoE). We implement and test our algorithm together with a number of state-of-the-art quality adaptation algorithms. Through extensive experiments on both testbed and cross-Atlantic Internet scenarios, we show that by sacrificing little to none in average quality bitrate, SQUAD provides significantly better QoE in terms of number and magnitude of quality switches.