A Hierarchical Characterization of a Live Streaming Media Workload

We present what we believe to be the first thorough characterization of live streaming media content delivered over the Internet. Our characterization of over five million requests spanning a 28-day period is done at three increasingly granular levels, corresponding to clients, sessions, and transfers. Our findings support two important conclusions. First, we show that the nature of interactions between users and objects is fundamentally different for live versus stored objects. Access to stored objects is user driven, whereas access to live objects is object driven. This reversal of active/passive roles of users and objects leads to interesting dualities. For instance, our analysis underscores a Zipf-like profile for user interest in a given object, which is to be contrasted to the classic Zipf-like popularity of objects for a given user. Also, our analysis reveals that transfer lengths are highly variable and that this variability is due to the stickiness of clients to a particular live object, as opposed to structural (size) properties of objects. Second, based on observations we make, we conjecture that the particular characteristics of live media access workloads are likely to be highly dependent on the nature of the live content being accessed. In our study, this dependence is clear from the strong temporal correlations we observed in the traces, which we attribute to the synchronizing impact of live content on access characteristics. Based on our analyses, we present a model for live media workload generation that incorporates many of our findings, and which we implement in GISMO [19].

Cache-and-Relay Streaming Media Delivery for Asynchronous Clients

We consider the problem of delivering popular streaming media to a large number of asynchronous clients. We propose and evaluate a cache-and-relay end-system multicast approach, whereby a client joining a multicast session caches the stream, and if needed, relays that stream to neighboring clients which may join the multicast session at some later time. This cache-and-relay approach is fully distributed, scalable, and efficient in terms of network link cost. In this paper we analytically derive bounds on the network link cost of our cache-and-relay approach, and we evaluate its performance under assumptions of limited client bandwidth and limited client cache capacity. When client bandwidth is limited, we show that although finding an optimal solution is NP-hard, a simple greedy algorithm performs surprisingly well in that it incurs network link costs that are very close to a theoretical lower bound. When client cache capacity is limited, we show that our cache-and-relay approach can still significantly reduce network link cost. We have evaluated our cache-and-relay approach using simulations over large, synthetic random networks, power-law degree networks, and small-world networks, as well as over large real router-level Internet maps.

Scalable Overlay Multicast Tree Construction for QoS-Constrained Media Streaming

Overlay networks have become popular in recent times for content distribution and end-system multicasting of media streams. In the latter case, the motivation is based on the lack of widespread deployment of IP multicast and the ability to perform end-host processing. However, constructing routes between various end-hosts, so that data can be streamed from content publishers to many thousands of subscribers, each having their own QoS constraints, is still a challenging problem. First, any routes between end-hosts using trees built on top of overlay networks can increase stress on the underlying physical network, due to multiple instances of the same data traversing a given physical link. Second, because overlay routes between end-hosts may traverse physical network links more than once, they increase the end-to-end latency compared to IP-level routing. Third, algorithms for constructing efficient, large-scale trees that reduce link stress and latency are typically more complex. This paper therefore compares various methods to construct multicast trees between end-systems, that vary in terms of implementation costs and their ability to support per-subscriber QoS constraints. We describe several algorithms that make trade-offs between algorithmic complexity, physical link stress and latency. While no algorithm is best in all three cases we show how it is possible to efficiently build trees for several thousand subscribers with latencies within a factor of two of the optimal, and link stresses comparable to, or better than, existing technologies.

Efficient delivery of scalable media streaming over lossy networks

Recent years have witnessed a rapid growth in the demand for streaming video over the Internet, exposing challenges in coping with heterogeneous device capabilities and varying network throughput. When we couple this rise in streaming with the growing number of portable devices (smart phones, tablets, laptops) we see an ever-increasing demand for high-definition videos online while on the move. Wireless networks are inherently characterised by restricted shared bandwidth and relatively high error loss rates, thus presenting a challenge for the efficient delivery of high quality video. Additionally, mobile devices can support/demand a range of video resolutions and qualities. This demand for mobile streaming highlights the need for adaptive video streaming schemes that can adjust to available bandwidth and heterogeneity, and can provide us with graceful changes in video quality, all while respecting our viewing satisfaction. In this context the use of well-known scalable media streaming techniques, commonly known as scalable coding, is an attractive solution and the focus of this thesis. In this thesis we investigate the transmission of existing scalable video models over a lossy network and determine how the variation in viewable quality is affected by packet loss. This work focuses on leveraging the benefits of scalable media, while reducing the effects of data loss on achievable video quality. The overall approach is focused on the strategic packetisation of the underlying scalable video and how to best utilise error resiliency to maximise viewable quality. In particular, we examine the manner in which scalable video is packetised for transmission over lossy networks and propose new techniques that reduce the impact of packet loss on scalable video by selectively choosing how to packetise the data and which data to transmit. We also exploit redundancy techniques, such as error resiliency, to enhance the stream quality by ensuring a smooth play-out with fewer changes in achievable video quality. The contributions of this thesis are in the creation of new segmentation and encapsulation techniques which increase the viewable quality of existing scalable models by fragmenting and re-allocating the video sub-streams based on user requirements, available bandwidth and variations in loss rates. We offer new packetisation techniques which reduce the effects of packet loss on viewable quality by leveraging the increase in the number of frames per group of pictures (GOP) and by providing equality of data in every packet transmitted per GOP. These provide novel mechanisms for packetizing and error resiliency, as well as providing new applications for existing techniques such as Interleaving and Priority Encoded Transmission. We also introduce three new scalable coding models, which offer a balance between transmission cost and the consistency of viewable quality.

SDC: Scalable description coding for adaptive streaming media

Video compression techniques enable adaptive media streaming over heterogeneous links to end-devices. Scalable Video Coding (SVC) and Multiple Description Coding (MDC) represent well-known techniques for video compression with distinct characteristics in terms of bandwidth efficiency and resiliency to packet loss. In this paper, we present Scalable Description Coding (SDC), a technique to compromise the tradeoff between bandwidth efficiency and error resiliency without sacrificing user-perceived quality. Additionally, we propose a scheme that combines network coding and SDC to further improve the error resiliency. SDC yields upwards of 25% bandwidth savings over MDC. Additionally, our scheme features higher quality for longer durations even at high packet loss rates.

A fast, streaming SIMD extensions 2, logistic squashing function

Schraudolph proposed an excellent exponential approximation providing increased performance particularly suited to the logistic squashing function used within many neural networking applications. This note applies Intel's streaming SIMD Extensions 2 (SSE2), where SIMD is single instruction multiple data, of the Pentum IV class processor to Schraudolph's technique, further increasing the performance of the logistic squashing function. It was found that the calculation of the new 32-bit SSE2 logistic squashing function described here was up to 38 times faster than the conventional exponential function and up to 16 times faster than a Schraudolph-style 32-bit method on an Intel Pentum D 3.6 GHz CPU.

A Kernel Interleaved Scheduling Method for Streaming Applications on Soft-core Vector Processors

Massively parallel networks of highly efficient, high performance Single Instruction Multiple Data (SIMD) processors have been shown to enable FPGA-based implementation of real-time signal processing applications with performance and
cost comparable to dedicated hardware architectures. This is achieved by exploiting simple datapath units with deep processing pipelines. However, these architectures are highly susceptible to pipeline bubbles resulting from data and control hazards; the only way to mitigate against these is manual interleaving of
application tasks on each datapath, since no suitable automated interleaving approach exists. In this paper we describe a new automated integrated mapping/scheduling approach to map algorithm tasks to processors and a new low-complexity list scheduling technique to generate the interleaved schedules. When applied to a spatial Fixed-Complexity Sphere Decoding (FSD) detector
for next-generation Multiple-Input Multiple-Output (MIMO) systems, the resulting schedules achieve real-time performance for IEEE 802.11n systems on a network of 16-way SIMD processors on FPGA, enable better performance/complexity balance than current approaches and produce results comparable to handcrafted implementations.

Particle simulation study of electron heating by counter-streaming ion beams ahead of supernova remnant shocks

The growth and saturation of Buneman-type instabilities is examined with a particle-in-cell (PIC) simulation for parameters that are representative for the foreshock region of fast supernova remnant shocks. A dense ion beam and the electrons correspond to the upstream plasma and a fast ion beam to the shock-reflected ions. The purpose of the 2D simulation is to identify the nonlinear saturation mechanisms, the electron heating and potential secondary instabilities that arise from anisotropic electron heating and result in the growth of magnetic fields. We confirm that the instabilities between both ion beams and the electrons saturate by the formation of phase space holes by the beam-aligned modes. The slower oblique modes accelerate some electrons, but they cannot heat up the electrons significantly before they are trapped by the faster beam-aligned modes. Two circular electron velocity distributions develop, which are centred around the velocity of each ion beam. They develop due to the scattering of the electrons by the electrostatic wave potentials. The growth of magnetic fields is observed, but their amplitude remains low.