Measurement-based study on the relation between users’ watching behavior and network sharing in P2P VoD systems

P2P-based Video on Demand (P2P VoD) systems can provide higher quality video services than P2P live systems. However, owing to the unavailability of open source codes, the characteristics of P2P VoD systems have not been fully explored. For this purpose, in this paper, we first implemented reverse engineering to the protocols of two most popular VoD systems used in China. Peers’ watching behavior and network sharing were then studied through the measured buffer messages. Based on our study, we found that uploading-only peers were more than downloading peers, and smooth-watching peers were more than randomly-seeking peers in the current P2P VoD systems. This finding is significantly different from the existing P2P-based file dissemination systems and the results obtained in the previous work on VoD systems. Furthermore, this study revealed the relations between watching behavior and network sharing. It has been observed that randomly-seeking peers will provide more regular as well as rare chunks than smooth-watching peers. A simple mathematical model was established to analytically demonstrate the relation between the watching index (WI) distribution and network-sharing profile, and this relation was also validated by our measured data.     

CircleSearch：流媒体VoD环境下的P2P搜索算法

在基于P2P的流媒体VoD环境下,搜索机制是影响服务质量的关键因素之一。该文提出一种新的基于环状网络的分布式自适应概率搜索算法——CircleSearch。它依据节点本体内容生成结构化的环状拓扑网络,保证算法的可扩展性、健壮性和分布性。自发组织基于本体距离和连接可靠性的small world覆盖网络,减小消息的转发次数,采用自适应搜索算法提高文件定位的准确性。仿真实验表明,CircleSearch以其高成功率、低服务器负载和很小的平均搜索长度实现媒体流传输的即时、稳定、可控和连续,为VoD服务提供QoS保证。     

一种用于大规模P2P点播系统的拓扑结构

提出了一种基于动态时间坐标的分层DHT拓扑结构,解决了因大规模P2P点播系统要求细粒度追踪而难以应用DHT的问题。在动态时间坐标系中,节点的播放点坐标不再随着节点的播放而移动,从而使得DHT能够用于追踪点播系统节点缓存位置。仿真结果证明了方法的有效性。     

SmartPeerCast: a Smart QoS driven P2P live streaming framework

The P2P swarm technologies have been shown to be very efficient for medium scale content distribution systems in the last few years, such as the file sharing and video-on-demand (VOD) applications. However it is still an open topic about how to deploy the P2P paradigm for the real time video broadcasting (RTVB) applications. The P2P RTVB application is different from the cache based P2P system because it has more stringent restrictions for startup time and packet loss ratio. In this paper, an adaptive media broadcasting P2P framework named SmartPeerCast which employs the media transrating service to control the quality of service (QoS), is proposed. SmartPeerCast achieves a network awareness, codec awareness, and high performance RTVB service with four key designs: (1) It groups the newly joined peers into different quality clusters by their uploading capability. This clustering mechanism avoids the bandwidth bottleneck between the heterogeneous peers of the overall P2P overlay by only forwarding the same quality stream over the peers in the same cluster. (2) The streaming quality is adjusted adaptively between the sending and the receiving peers by a Smart QoS algorithm to compensate for the network jitters to reduce the receiving peer’s playback jitter. (3) The receiving peer monitors the data forwarding QoS of the sending peer to select the best suitable parent node dynamically. The SmartPeerCast uses this Smart QoS framework to implement an incentive mechanism to award the peers with high uploading contributions by migrating them to a higher quality cluster. (4) A transrating engine is used at the leaf nodes of the high quality cluster to forward the stream with suitable bits rate to the nodes of the low quality cluster; this transrating service not only can fully utilize the uploading bandwidth of the peers in the higher quality cluster but also avoids the bandwidth bottleneck of stream forwarding between the heterogeneous peers. Our experiment results and the real deployment show that SmartPeerCast can eliminate the bandwidth bottleneck and content bottleneck between the heterogeneous peers with a smaller startup time and packet loss and it is a high performance and medium scale P2P RTVB framework.     

Analysis of distribution time of multiple files in a P2P network

Peer-to-Peer (P2P) file sharing attracts much attention due to its scalability and robustness. One important metric in measuring the performance of a P2P file sharing system is the amount of time required for all peers to get the files. We refer this time as the file distribution time. Researchers have proposed protocols to minimize the file distribution time under different situation. However, most works are based on the single-file scenario. On the other hand, there are studies showing that in a file sharing application, users may download multiple files at the same time. In this paper, we analyze the minimum time needed to distribute multiple files. We develop an explicit expression for the minimum amount of time needed to distribute multiple files in a heterogeneous P2P fluid model. Unlike the single-file scenario, we demonstrate that the theoretical lower bound in multi-file is not always achievable. With a comprehensive consideration of all the configurations, we elaborate how to partition the bandwidth capacities of both seeds and leechers for a particular file such that the finish time is optimal.     

P2P网络下一种基于DHT的视频点播方案

在基于P2P的视频点播系统中,节点邻居选择策略对服务质量有很大的影响。提出一种基于DHT（Distributed Hash Table）的P2P覆盖网络下视频点播（Video-on—demand）的解决方案。通过网络坐标系统的拓扑发现能力,充分结合DHT网络高速搜索和VoD视频点播的特性、构造一种具有高效邻居选择能力、高用户自由度、高可靠性、扩展性的体系架构。针对架构设计中资源发布份发、资源的搜索、视频点播的实现等关键问题提出了解决方案,分析了系统的特性。     

A Trace-Driven Approach to Evaluate the Scalability of P2P-Based Video-on-Demand Service

Peer-to-Peer (P2P) networks have emerged as one of the most promising approaches to improve the scalability of Video-on-Demand (VoD) service over Internet. However, despite a number of architectures and streaming protocols have been proposed in past years, there is few work to study the practical performance of P2P-based VoD service especially in consideration of real user behavior which actually has significant impact on system scalability. Therefore, in this paper, we first characterize the user behavior by analyzing a large amount of real traces from a popular VoD system supported by the biggest television station in China, cctv.com. Then we ex-amine the practical scalability of P2P-based VoD service through extensive trace-driven simula-tion under a general system framework. The results show that P2P networks scale well in provid-ing VoD service under real user behavior by obtaining a considerable good cache hit ratio. Moreover, it is observed that adopting hard cache at client side help achieves better system scal-ability than that with soft cache. We also identify the impact of various aspects of user behavior upon system scalability through detailed simulation. We believe that our study will shine insight-ful light on the understanding of practical scalability of P2P-based VoD service and be helpful to future system design and optimization.     

Modeling and analysis of multi-channel P2P VoD systems

Peer-to-Peer (P2P) technology has recently become a tremendously attractive solution to offload servers in large-scale Video-on-Demand (VoD) applications by utilizing the upload capability of participating peers. However, the alleviation of server load may not be very effective in the multi-channel P2P VoD systems due to the intra-channel and inter-channel upload bandwidth imbalances. In this paper, we evaluate the impact of these bandwidth imbalances on the system performance through two analytically models: a queuing network model which describes the whole system, and a stage-based model which describes a single channel. Our models capture several aspects of peer behavior, such as participating in the system, sojourning in a channel, downloading and uploading the content, wandering around channels and leaving the system. We apply our models to three potential P2P VoD designs: Naive bandwidth Allocation Design (NAD), Independent-channel Chunk-aware bandwidth Allocation design (ICA) and Cross-channel Chunk-aware bandwidth Allocation design (CCA). By developing an asymptotic theory to provide theoretical results, we analytically show that ICA can perform better than NAD, and CCA can perform best of the three designs in terms of server load. We also present optimization problems and simple heuristic strategies for ICA and CCA. Our analytical results are validated by extensive simulations.     

动态段阈值机制在P2Cast VoD系统中的应用

本文对一种新型的视频点播系统P2Cast视频点播系统的关键技术进行了研究,针对段阈值在系统的可扩展性方面以及视频服务质量方面所起的重要作用,结合补丁流算法和客户节点的请求速率提出了一种嵌入动态调节段闲值机制的BF改进算法。该算法在一定程度上提高了P2Cast视频点播系统的性能。     

Investigation of a large-scale P2P VoD overlay network by measurements

The proliferation of large-scale Peer-to-Peer (P2P) Video on Demand (VoD) overlays has created the demands to characterize and to understand the emergent properties of P2P VoD applications. In this paper, we investigate the P2P VoD overlay network, spanned by PPLive application, using active measurements. We have developed a piece of measurement software, named as VoDCrawler, to implement the experiment. Different from other existing work, our study focuses on the topology and the churn of the overlay networks. From the study, we have revealed the major features of the P2P VoD overlay networks and have compared them with those in P2P file sharing and live streaming systems. The discoveries we have obtained could help further investigation of the operation of the P2P VoD systems and favor the commercial developments of the P2P VoD applications.     

基于P2P的流媒体点播技术研究与展望

视频点播数据传输的离散程度,给P2P流媒体点播技术带来了极大的挑战.分析了P2P流媒体点播技术的基本问题,从媒体资源定位和分发两个方面介绍各种已提出的协议和算法,结合P2P视频点播的特点和网络应用的实际需求分析其优点和不足,并以此为基础提出有待解决的主要问题,为进一步的研究和应用提供有益的参考.     

一种基于节点可选度的P2P点播流媒体数据调度算法

当前的P2P(Peer-to-Peer)点播流媒体系统中数据调度算法未能充分利用每个用户节点自身的特性.在分析典型数据调度算法基础上提出一种基于节点可选度的数据调度算法(SSP算法).该算法一方面在调度下载数据块时综合考虑了邻居节点带宽能力及其所拥有的数据信息.另一方面对服务节点的请求处理过程进行了优化.SSP算法有利于提高用户节点播放视频的连续性,降低流媒体服务器的负载压力,从而改善P2P点播流媒体系统的整体服务质量.仿真结果和实际应用表明算法性能良好,适用于用户节点能力差异较大的P2P点播流媒体环境.     

面向跳播优化的P2P视频点播系统

基于P2VoD算法,针对其不足之处进行改变和优化,设计了新的系统。新的系统以树网结合的拓扑结构,实现对等体的加入、跳播、离开等行为的处理,尤其是对点播用户跳播行为的支持进行了特别的优化,减小了系统的延迟抖动。经过仿真,结果表明在相同的网络负载下,对等体初始加入网络时的延迟比P2VoD要小20％,同时系统比P2VoD更加稳定,并且实现了对跳播功能的支持,平均跳播时延约为初始时延的50％。     

对等网络中不同信任模型的比较分析

在诸如文件共享等的Peer-to-Peer环境中,资源共享是用户自愿的行为,在这类系统中由于用户不必为自身的行为担负(法律)责任,因而节点间的信任关系就显得尤为重要。对几种信任模型进行比较,这些模型分别侧重于评估各个节点的服务质量和惩罚不信任的节点等不同侧面,经过比较分析,得出建立P2P信任模型需要考虑的几个方面。     

Measurement,modeling and enhancement of BitTorrent-based VoD system

BitTorrent is one of the most popular Peer-to-Peer (P2P) applications for file sharing over the Internet. Video files take up a large proportion of space among the BitTorrent shared files. Recently, BitTorrent has attracted researchers’ interests, as an alternative method of providing video on demand (VoD) service. In this paper, we concentrate on enabling BitTorrent to support VoD service in existing swarms while maintaining the download efficiency of file-sharing users. We first examine the content properties of the BitTorrent system to explore the demands and challenges of VoD service in BitTorrent swarms. The efficiency of BitTorrent for various piece selection policies is then compared through measurement on PlanetLab. We also use an optimization mathematical model to analyze the hybrid BitTorrent system in which downloading peers and streaming peers coexist. Both measurement results and model analysis indicate the problem of system efficiency decline in the BitTorrent-based VoD systems, in comparison with the original BitTorrent file-sharing system. Our proposed approach, unlike existing strategies that are limited to changing the piece selection policy to allow BitTorrent to support streaming services, modifies both piece and peer selection policies to provide a “streaming while downloading” service in the BitTorrent system with downloading peers. For the peer selection policy, a CAP (Closest-Ahead Peers) method is applied to make better use of the peers’ upload bandwidths. For the piece selection policy, a sliding window-based hybrid method that combines the rarest-first policy with the sequential policy is proposed. To demonstrate the performance of our proposed approach, an evaluation is made using various metrics on PlanetLab. The results show that our proposed method has higher throughput and better streaming continuity than the sequential policy and BiToS.     

On incentivizing upload capacity in P2P-VoD systems: Design,analysis and evaluation

Free riding has long been a serious problem in peer-to-peer (P2P) systems due to the selfish behavior of individual users. To conquer this problem, a key design issue of the P2P systems is to appropriately incentivize users to contribute resources. In P2P Video-on-Demand (VoD) applications, content providers need to incentivize the peers to dedicate bandwidth and upload data to one other so as to alleviate the upload workload of their content servers. In this paper, we design a simple yet practical incentive mechanism that rewards each peer based on its dedicated upload bandwidth. We use a mean field interaction model to characterize the distribution of number of peers in different video segments, based on which we characterize the content providers’ uploading cost as a function of the peers’ contribution. By using a game theoretic framework, we analyze the interaction between a content provider’s rewarding strategy and the peers’ contributing behaviors and derive a unique Stackelberg equilibrium. We further analyze the system efficiency in terms of the price of anarchy and study the long term behavior of the system under a repeated game setting. Via extensive simulations, we validate the stability and efficiency of our incentive scheme.     

Measurement study on P2P streaming systems

Originally used as the default infrastructure for efficient file sharing, peer-to-peer (P2P) architecture achieved great successes. Now, the P2P model has been adopted for many other distributed applications, such as instant message and phone services, Internet gaming, and large-scale scientific computing. In recent years, P2P streaming systems experienced tremendous growth and became one of the largest bandwidth consumers on the Internet. Compared to standard file sharing systems, the streaming services show unique characteristics with more stringent time constraints and require much higher network bandwidth. It is extremely important to evaluate and analyze existing applications, and investigate the merits and weaknesses in these systems for future development. In this paper, we conduct a comprehensive measurement study on two of the most popular P2P streaming systems, namely, PPLive and PPStream. They are very popular P2P streaming applications, and serving millions of registered users with hundreds of live TV channels and millions of other video clips. In our measurement, we deploy our collectors in China, and both live TV and video-on-demand (VoD) channels are evaluated. We record run-time network traffic on the client side, compare and analyze the characteristics of these channels based on their popularity. For both categories, we perceive that, in general, the two measured P2P streaming systems provide satisfactory experience to the audiences for all channels regardless of popularity. However, the most of data are downloaded from the dedicated servers for unpopular channels. We also observe that live TV channels show better peer coordination than VoD channels. Beside the traffic, we have also collected cache replacement information for VoD channels, and these measurement results can help us understand the caching mechanism of P2P streaming systems. With the support of the cache, VoD channels perform better than their counterparts in the live TV category in terms of data transmission, workload distribution, and signal traffic overhead. Overall, our results reveal that although P2P streaming systems can usually provide excellent viewing experience for popular channels, there are still challenges to fully support unpopular channels. New designs and algorithms are in urgent need, especially for unpopular live TV channels.     

P2P视频点播内容分发策略

视频点播目前已成为对等(peer-to-peer,简称P2P)网络中一项重要的应用,引起了人们的不少研究兴趣.由于P2P网络能够为VoD(video-on-demand)应用的大规模实现提供底层网络的支持,许多正在出现的P2P VoD分发策略都能够提供在P2P网络中最基本的数据传输方式.对以往主要的P2P VoD内容分发策略进行了总结和概括.首先介绍了设计P2P VoD策略的相关重要问题,并把策略根据内容分发方式的不同分成4种类型.最后讨论了它们的应用层性能,并提出未来可以延续的工作.     

Distributed Throughput Maximization in P2P VoD Applications

In peer-to-peer (P2P) video-on-demand (VoD) systems, a scalable source coding is a promising solution to provide heterogeneous peers with different video quality. In this paper, we present a systematic study on the throughput maximization problem in P2P VoD applications. We apply network coding to scalable P2P systems to eliminate the delivery redundancy. Since each peer receives distinct packets, a peer with a higher throughput can reconstruct the video at a higher quality. We maximize the throughput in the existing buffer-forwarding P2P VoD systems using a fully distributed algorithm. We demonstrate in the simulations that the proposed distributed algorithm achieves a higher throughput compared to the proportional allocation scheme or the equal allocation scheme. The existing buffer-forwarding architecture has a limitation in total upload capacity. Therefore we propose a hybrid-forwarding P2P VoD architecture to improve the throughput by combining the buffer-forwarding approach with the storage-forwarding approach. The throughput maximization problem in the hybrid-forwarding architecture is also solved using a fully distributed algorithm. We demonstrate that the proposed hybrid-forwarding architecture greatly improves the throughput compared to the existing buffer-forwarding architecture. In addition, by adjusting the priority weight at each peer, we can implement the differentiated throughput among different users within a video session in the buffer-forwarding architecture, and the differentiated throughput among different video sessions in the hybrid-forwarding architecture.      

Zebroid: using IPTV data to support STB-assisted VoD content delivery

IPTV, unlike Internet TV, delivers digital TV and multimedia services over IP-based networks with the required level of quality of service (QoS) and quality of experience (QoE). Linear programming channels in IPTV are delivered through multicast, which is highly scalable with the number of subscribers. Video-on-demand (VoD) content, on the other hand, is typically delivered using unicast, which places a heavy load on the VoD servers and all the network components leading to the end-user set-top boxes (STBs) as the demand increases. With the rapid growth of IPTV subscribers and the shift in video viewing habits, the need to efficiently disseminate large volumes of VoD content has prompted IPTV service providers to consider the use of STBs to assist in video content delivery. This paper describes our current research work on Zebroid, a potential VoD solution for fiber-to-the-node (FTTN) networks, which uses IPTV data on a recurring basis to determine how to select, stripe, and preposition popular content in selected STBs during idle hours. A STB requesting VoD content during the peak hours can then receive necessary stripes from participating STBs in the neighborhood. Recent VoD request access patterns, STB availability data, and capacity data on network components are taken into consideration in determining the parameters used in the striping algorithm of Zebroid. We show both by simulation and emulation on a realistic IPTV testbed that the VoD server load can be reduced by more than 70% during peak hours by allocating only 8 GB of storage on each STB. The savings achieved through Zebroid would also allow IPTV service providers to add more linear programming channels without expensive infrastructure upgrades.