... to flood Internet access and backbone ISPs with massive amounts of new traffic. We recently measured 200,000 IPTV users for a single program, receiving at an aggregate simultaneous rate of 100 gigabits/second. Although many architectures are possible for IPTV video distribution, several chunkdriven P2P architectures have been successfully deployed in the Internet. In order to gain insight into chunk-driven P2P IPTV systems and the traffic loads they place on ISPs, we have undertaken an in-depth measurement study of one of the most popular IPTV systems, namely, PPLive. We have developed a dedicated PPLive crawler, which enables us to study the global characteristics of the chunk-driven PPLive system. We have also collected extensive packet traces for various different measurement scenarios, including both campus access network and residential access networks. The measurement results obtained through these platforms bring important insights into IPTV user behavior, P2P IPTV traffic overhead and redundancy, peer partnership characteristics, P2P IPTV viewing quality, and P2P IPTV design principles.

Abstract—We develop a simple stochastic fluid model that seeks to expose the fundamental characteristics and limitations of P2P streaming systems. This model accounts for many of the essential features of a P2P streaming system, including the peers ’ realtime demand for content, peer churn (peers joining and leaving), peers with heterogeneous upload capacity, limited infrastructure capacity, and peer buffering and playback delay. The model is tractable, providing closed-form expressions which can be used to shed insight on the fundamental behavior of P2P streaming systems. The model shows that performance is largely determined by a critical value. When the system is of moderate-to-large size, if a certain ratio of traffic loads exceeds the critical value, the system performs well; otherwise, the system performs poorly. Furthermore, large systems have better performance than small systems since they are more resilient to bandwidth fluctuations caused by peer churn. Finally, buffering can dramatically improve performance in the critical region, for both small and large systems. In particular, buffering can bring more improvement than can additional infrastructure bandwidth. I.

In recent years, overlay networks have become an important vehicle for delivering Internet applications. Overlay network nodes are typically implemented using general purpose servers or clusters. We investigate the performance benefits of more integrated architectures, combining general-purpose servers with high performance Network Processor (NP) subsystems. We focus on PlanetLab as our experimental context and report on the design and evaluation of an experimental PlanetLab platform capable of much higher levels of performance than typical system configurations. To make it easier for users to port applications, the system supports a fast path/slow path application structure that facilitates the mapping of the most performance-critical parts of an application onto an NP subsystem, while allowing the more complex control and exception-handling to be implemented within the programmer-friendly environment provided by conventional servers. We report on implementations of two sample applications, an IPv4 router, and a forwarding application for the Internet Indirection Infrastructure. We demonstrate an 80 × improvement in packet processing rates and comparable reductions in latency. This work supported in part by NSF (grants 0520778 and 0626661).

Global-scale Content Distribution Networks (CDNs), such as Akamai, distribute thousands of servers worldwide providing a highly reliable service to their customers. Not only has reliability been one of the main design goals for such systems — they are engineered to operate under severe and constantly changing number of server failures occurring at all times. Consequently, in addition to being resilient to component or network outages, CDNs are inherently considered resilient to denial-of-service (DoS) attacks as well. In this paper, we focus on Akamai’s (audio and video) streaming service and demonstrate that the current system design is highly vulnerable to intentional service degradations. We show that (i) the discrepancy among streaming flows ’ lifetimes and DNS redirection timescales, (ii) the lack of isolation among customers and services, (e.g., video on demand vs. live streaming), (iii) a highly transparent system design, (iv) a strong bias in the stream popularity, and (v) minimal clients ’ tolerance for low-quality viewing experiences, are all factors that make intentional service degradations highly feasible. We demonstrate that it is possible to impact arbitrary customers ’ streams in arbitrary network regions: not only by targeting appropriate points at the streaming network’s edge, but by effectively provoking resource bottlenecks at a much higher level in Akamai’s multicast hierarchy. We provide countermeasures to help avoid such vulnerabilities and discuss how lessons learned from this research could be applied to improve DoS-resiliency of large-scale distributed and networked systems in general.

Abstract. IP networks present a challenging environment for video streaming because they do not provide throughput, jitter, or loss rate guarantees. In this work, we focus on improving the perceived quality of video streaming through dynamic path selection. Selecting one of several Internet paths is possible using multihoming and/or an overlay routing infrastructure. We conduct an experimental comparison of various measurement-based path selection techniques for video streaming. The path selection is based on the measurement of network-layer metrics, such as loss rate, jitter or available bandwidth, while the video quality is evaluated based on the VQM tool. Our experiments show that the most effective technique for adaptive path selection relies on an estimate of the lower bound of the available bandwidth variation range. We show how to perform such measurements using the video packets, eliminating the measurement overhead in the selected path. Finally, we show that adaptive path selection is more effective than a simple, but commonly used, form of FEC. 1

Video-on-demand (VoD) service providers are intensely interested in transport, storage, streaming and caching in content delivery networks. Today’s 5,000-hour library may grow toward the 750,000-hour “Long Tail ” movie and TVseries catalog. We propose a method to calculate how much of a library should be cached. Much previous work focused on theoretical caching concepts, or the dynamics of cache filling and reclamation. Our method explicitly considers the impact of the available video server equipment; we present a VoD design tool comprising a novel cost function, hit ratio estimation and heuristic. 1.

Deploying and managing wide-area network services is exceptionally challenging. Despite having servers at many locations, a service provider must rely on an underlying besteffort network; a network provider can offer services over its own customized network, but only within limited footprint. In this paper, we propose Cabernet (Connectivity Architecture for Better Network Services), a three-layer network architecture that lowers the barrier for deploying wide-area services. We introduce the connectivity layer, which uses virtual links purchased from infrastructure providers to run virtual networks with the necessary geographic footprint, reliability, and performance for the service providers. As an example, we present a cost-effective way to support IPTV delivery through wide-area IP multicast that runs on top of a reliable virtual network. 1.

Abstract — The paper reports on recent developments and challenges in reliable multicast communication, with special focus on reliable multicast communication at the application layer. The foundation of reliable multicast communication is given by several components, which are multicast communication, congestion control and error control. Our paper is providing a survey of these mechanisms in multicast environments. I.

Overlay hosting systems such as PlanetLab, and cloud computing environments such as Amazon’s EC2, provide shared infrastructures within which new applications can be developed and deployed on a global scale. This paper explores how systems of this sort can be used to enable advanced network services and sophisticated applications that use those services to enhance performance and provide a high quality user experience. Specifically, we investigate how advanced overlay hosting environments can be used to provide network services that enable scalable virtual world applications and other large-scale distributed applications requiring consistent, real-time performance. We propose a novel network architecture called Forest built around persession tree-structured communication channels that we call comtrees. Comtrees are provisioned and support both unicast and multicast packet delivery. The multicast mechanism is designed to be highly scalable and lightweight enough to support the rapid changes to multicast subscriptions needed for efficient support of state updates within virtual worlds. We evaluate performance using a combination of analysis and experimental measurement of a partial system prototype that supports fully functional distributed game sessions. Our results provide the data needed to enable accurate projections of performance for a variety of session and system configurations.

Abstract — We develop a simple stochastic fluid model that seeks to expose the fundamental characteristics and limitations of P2P streaming systems. This model accounts for many of the essential features of a P2P streaming system, including the peers ’ real-time demand for content, peer churn (peers joining and leaving), peers with heterogeneous upload capacity, limited infrastructure capacity, and peer buffering and playback delay. The model is tractable, providing closed-form expressions which can be used to shed insight on the fundamental behavior of P2P streaming systems. The model shows that performance is largely determined by a critical value. When the system is of moderate-to-large size, if a certain ratio of traffic loads exceeds the critical value, the system performs well; otherwise, the system performs poorly. Furthermore, large systems have better performance than small systems since they are more resilient to bandwidth fluctuations caused by peer churn. Finally, buffering can dramatically improve performance in the critical region, for both small and large systems. In particular, buffering can bring more improvement than can additional infrastructure bandwidth. I.