P2P网络的实时性问题*

由于P2P网络的动态性、分散性和半可信性,其实时性研究和应用成为公认的难题。当前P2P多媒体应用系统和实时计算系统或多或少采用了实时保障措施,但均没有对P2P实时性进行系统研究。在总结现有P2P网络系统采用的实时技术的基础上,分析了P2P网络结构、搜索协议（算法）和任务调度对实时性的影响,提出P2P网络实时性研究存在的问题,总结出P2P实时性研究需要解决的关键技术。     

On peer-to-peer (P2P) content delivery

In both academia and industry, peer-to-peer (P2P) applications have attracted great attentions. P2P applications such as Napster, Gnutella, FastTrack, BitTorrent, Skype and PPLive, have witnessed tremendous success among the end users. Unlike a client-server based system, peers bring with them serving capacity. Therefore, as the demand of a P2P system grows, the capacity of the network grows, too. This enables a P2P application to be cheap to build and superb in scalability. In this paper, we survey the state of the art of the research and the development of P2P content delivery application. Using examples of the deployed P2P applications and research prototypes, we survey the best practices in P2P overlay building and P2P scheduling. We hope that the information may help the readers to build a reliable, robust P2P content delivery application.

一种综合的P2P网络信任模型

安全有效的信任模型是保证P2P 系统高效、稳定的关键技术之一。介绍一种适用于P2P 网络的综合信任模型,它参考了社会网络中信任关系的建立方法,从两个角度来计算节点的综合可信度。该模型中引入了非对称加密等安全机制,通过分析与仿真证明该模型能有效地抑制冒名和诋毁等非法行为,具有较强的安全性。     

Query-driven indexing for scalable peer-to-peer text retrieval

In this paper, we present a query-driven indexing/retrieval strategy for efficient full text retrieval from large document collections distributed within a structured P2P network. Our indexing strategy is based on two important properties: (1) the generated distributed index stores posting lists for carefully chosen indexing term combinations that are frequently present in user queries, and (2) the posting lists containing too many document references are truncated to a bounded number of their top-ranked elements. These two properties guarantee acceptable latency and bandwidth requirements, essentially because the number of indexing term combinations remains scalable and the posting lists transmitted during retrieval never exceed a constant size. A novel index update mechanism efficiently handles adding of new documents to the document collection. Thus, the generated distributed index corresponds to a constantly evolving query-driven indexing structure that efficiently follows current information needs of the users and changes in the document collection.We show that the size of the index and the generated indexing/retrieval traffic remains manageable even for Web-size document collections at the price of a marginal loss in precision for rare queries. Our theoretical analysis and experimental results provide convincing evidence about the feasibility of the query-driven indexing strategy for large scale P2P text retrieval.     

rlncBT：随机线性网络编码的P2P内容分发系统

基于线性编码和BT的原理和技术,设计和实现了一个基于随机线性网络编码的P2P内容分发系统,并对Piece选择策略进行了改进.通过仿真实验和理论分析,表明该系统具备下载成功率更高、平均下载时间更短、更好的健壮性、对网络带宽资源与服务器资源的利用率更高、对种子节点和服务器的依赖性更弱等多项优点。     

对等网联下NAT穿越问题的研究

采用"打洞"穿越锥型NAT与使用端口预测穿越对称型NAT相结合的方法,成功穿越了各种NAT,解决了对等联网下NAT的穿越问题。该方案既无需改变现有网络设备,又能确保内网的安全性,同时还解决了目前穿越方式存在的只能穿越部分类型的NAT、丢包、延时的问题。     

The method to secure scalability and high density in cloud data-center

Recently IT infrastructures change to cloud computing, the demand of cloud data center increased. Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared computing resources that can be rapidly provisioned and released with minimal management effort, the interest on data centers to provide the cloud computing services is increasing economically and variably. This study analyzes the factors to improve the power efficiency while securing scalability of data centers and presents the considerations for cloud data center construction in terms of power distribution method, power density per rack and expansion unit separately. The result of this study may be used for making rational decisions concerning the power input, voltage transformation and unit of expansion when constructing a cloud data center or migrating an existing data center to a cloud data center.     

Improving robustness of P2P applications in mobile environments

Mobile P2P networks possess particular characteristics which make accessibility of services deployed on peers a challenge. This has to be taken into account when considering robustness of applications that depend on successfully accessing a set of services. While ensuring robustness is traditionally handled through replication or redundancy, those solutions are not readily applicable to decentralized and dynamic networks. Instead, current solutions are based on efficient P2P structure maintenance or unstructured network search algorithms. A novel and alternative method proposed in this paper is based on the observation that some redundancy may exist between services offered on the network, a fact which could be used to recreate an unavailable service from services accessible to a peer. Instead of adding redundancy to the system, our solution exploits the already existing redundancy to improve robustness of mobile P2P applications. We model the interaction with services as finite-state transducers and propose a heuristic to obtain redundancy between any pair of services. Then, a set of algorithms that uses this inter-service redundancy to recreate the interaction with one service from the other is discussed. The computational cost is polynomial with respect to services’ size, and in practice, the non-redundant functionality and related control need to be implemented locally.

A novel DRM framework for peer-to-peer music content delivery

Due to rapid advances in the network communications field in recent years, the distribution of large-scale music contents has become easier and more efficient than ever before. However, the unauthorized distribution of copyright-protected content has emerged as a major concern. Accordingly, this paper presents a content distribution framework with a DRM capability for P2P networks. The robustness of the content distribution is ensured by using a network coding approach based on the Lagrange polynomial interpolation method. When the downloading peer within the network receives sufficient coded pieces, it not only reconstructs the associated blocks using a finite field Gaussian elimination method, but also creates its own copies of the coded pieces within these blocks and shares these copies amongst the other peers in the network. As a result, the distribution overhead imposed on the music provider is substantially reduced and the number of coded pieces within the network is significantly increased, thereby overcoming the “last piece problem” inherent in existing P2P schemes. In the DRM module of the framework, the RSA public-key cryptosystem is used to generate a unique digital fingerprint for every user within the network. The fingerprint is embedded within the music file in a protected form such that the music provider can establish the identification of any user performing an unauthorized distribution of the file. The experimental results confirm that the proposed framework provides an efficient and secure means of distributing large-scale copyright-protected music contents with no discernible degradation in the audio quality.     

基于CDN和P2P技术的流媒体系统设计

分析基于CDN和P2P技术的流媒体系统设计方案,指出它们的优缺点,给出了一种结合CDN和P2P技术的流媒体系统设计方案,提出了CDN内容路由的负载均衡算法,对P2P技术在CDN网络中的应用进行了研究,提出一种应用于CDN边缘节点P2P网络的拓扑结构并给出了节点失效恢复策略.     

结合CDN与P2P技术的混合流媒体系统研究

对基于CDN和P2P的流媒体系统进行了分析对比,指出了它们各自的优缺点,并根据因特网的结构特点,给出了一种结合这两种技术的混合流媒体系统的设计方案,提出了客户机端的内容缓存替换问题并给出了一种替换算法,对客户机离线的随机性与频繁性问题,文中给出了一种新的失效服务节点的快速替换查找策略。     

移动P2P网络超级节点选取机制研究

针对半分布式结构的移动P2P网络超级节点选取问题,提出一种新颖的超级节点选取机制,全面分析影响超级节点选取的各种因素,按着其特点和类型不同将其分为效益型属性和成本型属性,并在此基础上建立了超级节点选取的带约束多目标优化模型.采用新的约束处理方法,利用免疫克隆算法对超级节点选取问题进行求解.实验结果表明,该机制可以有效降低超级节点的失效率,提高系统查询效率,算法收敛速度快,有较好的可扩展性.     

Demaq/Transscale: Automated distribution and scalability for declarative applications

The goal of the Demaq/TransScale system is to automate the distribution of applications to multiple hosts. Today, in order to create highly scalable architectures, developers have to manually restructure the application logic, creating balanced application fragments and data partitions to distribute across the available host machines. Our approach automates much of this manual work. We show how a novel, messaging-based programming model allows to implement distribution as a source-level transformation that turns a non-distributed application specification into a set of programs that can be executed on the various machines of a cluster. The challenge is the identification of application fragments that can be executed independently of each other, such that each fragment can be executed using only local transactions. Further, our method allows developers to increase scalability (i.e. the number of possible fragments) by choosing from a library of automatically applicable rewrites. As a running example, and for the reported experimental results, we use TPC-App as a benchmark application.     

高性能可伸缩Web应用系统编程策略

性能和伸缩性是Web应用系统两个重要的质量属性,两者从不同方面反映了Web应用系统为客户提供响应时间的能力.创建高性能可伸缩的Web应用系统,以有效地减少系统更新换代的次数,满足企业不断增长的客户规模,并最大限度地保护企业投资,是一个复杂的系统工程,也是当前系统开发者亟待解决的一个问题,涉及系统的分析、设计、编码、部署等方面.基于J2EE平台,基于实践,给出创建高性能可伸缩Web应用系统的编程策略.     

IPTV与P2P的结合

针对IPTV所面临的技术瓶颈,阐述了IPTV与P2P技术结合的优势,并描述了相关工作流程。     

P2P和CDN中MDC流媒体的性能对比

文中检测并对比了基于多重描述编码（慨）的流媒体在对等网（P2P）和内容分发网络（CDN）中的性能。多个服务器同时为一个客户请求提供相同的描述，这样提高了网络传输的可靠性，并增加了服务器的数据传输率。用ns-2网络模拟器实现了这两种方法，实验结果表明：虽然P2P网络存在高度不稳定的情况，但是基于P2P的MDC流视频的质量明显比CDN上的好。     

覆盖网体系结构及应用研究

覆盖网以Internet为基础,通过构建虚拟的逻辑网络可以提高网络性能,灵活地部署新应用,是一种被广泛采用的网络技术。总结分析了目前两种有影响的覆盖网体系结构,覆盖网的几种典型应用,及各自的关键技术和研究热点;最后提出了覆盖网思想在移动网格中应用的探索,及未来的发展趋势。     

基于P2P和CDN的流媒体直播系统的设计与实现

对现有流媒体播放系统的相关技术进行了分析比较,指出了各自的优点以及存在的问题.结合CDN与P2P两种技术的优点,改善传统内容分发网络拓扑结构,把P2P的扩展能力和CDN的可靠性、可管理性有效地结合起来,设计并实现一种效率高、可扩展性好、稳定性强、可管理的流媒体直播系统.并且对系统的节点管理、缓存管理、数据调度等关键技术进行了介绍,最后对该系统的特性进行了分析.仿真实验表明,在大规模网络环境中,该流媒体直播系统比单纯的P2P系统在性能上有明显的提高.     

创建高性能与高伸缩性的J2EE应用

对于J2EE应用来说 ,性能和伸缩性问题是必须考虑的一个重要问题。将从J2EE应用程序体系结构出发 ,探究产生这些性能问题的根源 ,并提出一些原则来提高J2EE应用的性能和伸缩性。     

具有路由性能和蚁群优化的移动P2P路由策略*

目前的移动P2P网络路由策略不能较好适应网络拓扑结构的动态多变、网络和移动设备的资源有限等特点,以及不能较好解决路由建立和维护所带来的网络拥塞和资源消耗。针对上述问题,采用有限洪泛路由查询和移动agent路由查询相结合的策略,为每个移动节点提供丰富可靠、及时高效的路由信息。同时,使用改进的蚁群算法,综合考虑网络带宽、时延等多个路由性能指标,作为路由策略中路由选择机制。仿真研究证明,将所提出的理论与方法应用于移动P2P的路由选择和维护等问题,本算法在控制消息的开销、平均响应效率等方面具有良好的性能,对于网络