Evaluation of performance enhancing proxies in internet over satellite

Performance enhancing proxies (PEPs) are widely used to improve the performance of TCP over high delay‐bandwidth product links and links with high error probability. In this paper we analyse the performance of using TCP connection splitting in combination with web caching via traces obtained from a commercial satellite system. We examine the resulting performance gain under different scenarios, including the effect of caching, congestion, random loss and file sizes. We show, via analysing our measurements, that the performance gain from using splitting is highly sensitive to random losses and the number of simultaneous connections, and that such sensitivity is alleviated by caching. On the other hand, the use of a splitting proxy enhances the value of web caching in that cache hits result in much more significant performance improvement over cache misses when TCP splitting is used. We also compare the performance of using different versions of HTTP in such a system. Copyright © 2003 John Wiley & Sons, Ltd.     

The Content Driven Mobile Internet

The development of future mobile networks will be driven, in large part, by content and web based services. In this paper, we examine several performance, scalability and architectural challenges faced by future mobile web applications and how advanced mobile content delivery techniques can address these challenges. We review existing content delivery using a taxonomy that consists of three categories: network scaling, end system acceleration, and content and protocol optimization. While wireline content delivery focuses on network and server scalability, mobile content delivery will likely benefit most from optimizing radio link usage. We also present our ongoing work in this area, which extends the functionality of edge caching to the terminal, uses user interest correlation information to maintain low terminal power consumption and adds a new dimension to radio resource management. This revised version was published online in August 2006 with corrections to the Cover Date.     

基于媒体用户访问行为偏好模型的代理缓存算法

目前,代理缓存技术广泛应用于改善流媒体传输的服务质量.文章从实际用户日志文件的分析出发,利用发现的用户浏览流媒体对象时的行为分布模型,提出了一种新的视频流媒体缓存算法.仿真结果证明,该算法可以通过记录很少的用户访问信息获取较高的性能表现.     

Examining the effects of caching,compression and popup filtering on internet surfing

Over the years the number of Web users has increased dramatically unfortunately leading to the inherent problem of congestion. This can affect each user's surfing experience. This paper investigates download times associated with a web request, identifies where delays occur, and provides guidelines which can be followed by web developers to enable a faster and more efficient download and service for their users. We conclude that delays in Internet traffic and congestion can often be attributed to poorly developed sites (e.g. excessive image size). Copyright © 2005 John Wiley & Sons, Ltd.     

A survey on deploying mobile deep learning applications: A systemic and technical perspective

With the rapid development of mobile devices and deep learning, mobile smart applications using deep learning technology have sprung up. It satisfies multiple needs of users, network operators and service providers, and rapidly becomes a main research focus. In recent years, deep learning has achieved tremendous success in image processing, natural language processing, language analysis and other research fields. Despite the task performance has been greatly improved, the resources required to run these models have increased significantly. This poses a major challenge for deploying such applications on resource-restricted mobile devices. Mobile intelligence needs faster mobile processors, more storage space, smaller but more accurate models, and even the assistance of other network nodes. To help the readers establish a global concept of the entire research direction concisely, we classify the latest works in this field into two categories, which are local optimization on mobile devices and distributed optimization based on the computational position of machine learning tasks. We also list a few typical scenarios to make readers realize the importance and indispensability of mobile deep learning applications. Finally, we conjecture what the future may hold for deploying deep learning applications on mobile devices research, which may help to stimulate new ideas.     

Impacts of user-selfishness on cooperative content caching in social wireless networks

Cooperative caching can be an effective mechanism for reducing electronic content provisioning cost in Social Wireless Networks (SWNETs). These networks are formed by a collection of mobile data enabled devices physically gathering in settings such as university campus, malls, airport and other public places. In this paper, we first propose an optimal collaborative object caching policy in order to minimize the object provisioning cost in SWNETs with homogenous user requests and a peer-rebate model for promoting collaboration. Then using an analytical model we study the impacts of user selfishness on the provisioning cost and the earned rebate when certain nodes in an SWNET selfishly deviate from the optimal policy. User selfishness is motivated in order to either increase individually earned rebate or to reduce content provisioning cost. The analytical model is validated by experimental results from simulated SWNETs using the network simulator ns2.     

An altruistic cross‐layer recovering mechanism for ad hoc wireless networks

Video streaming services have restrictive delay and bandwidth constraints. Ad hoc networks represent a hostile environment for this kind of real‐time data transmission. Emerging mesh networks, where a backbone provides more topological stability, do not even assure a high quality of experience. In such scenario, mobility of terminal nodes causes link breakages until a new route is calculated. In the meanwhile, lost packets cause annoying video interruptions to the receiver. This paper proposes a new mechanism of recovering lost packets by means of caching overheard packets in neighbor nodes and retransmit them to destination. Moreover, an optimization is shown, which involves a video‐aware cache in order to recover full frames and prioritize more significant frames. Results show the improvement in reception, increasing the throughput as well as video quality, whereas larger video interruptions are considerably reduced. Copyright © 2014 John Wiley & Sons, Ltd.     

多控制器集群流媒体缓存代理的研究

为满足用户分散的大规模视频流媒体应用需求,该文提出了一种基于多控制器集群缓存代理结构(Multi- controller based Cluster Streaming Cache Proxy, MCSCP),该体系由一组控制器构成控制子系统,由多个内容存储器构成分布式存储子系统。对控制子系统的关键问题进行了深入研究:采用图论思想给出控制器组的选取算法,设计了一种基于优先级的主控制器选举和在线切换协议(Priority based Master-controller Election and Handover, PMEH)。建立数学模型分析了多控制器体系对系统可靠性的改善程度,讨论了控制器选取算法的性能,研究了协议中报文发送间隔变量与系统开销的关系,并使用仿真实验给出了该参数的最佳取值范围。     

一种有资源适应性的transcoding代理缓存机制

在基于transcoding代理的流媒体服务系统中,CPU和网络是两种潜在的瓶颈资源.本文提出了一种有资源适应性的transcoding代理缓存机制,统一考虑CPU和网络的资源需求,以提高系统的服务能力.首先推导了多版本缓存策略下网络收益和CPU收益的计算方法.通过引入一个时变的影响因子 α (t),给出了缓存系统聚合资源收益的表达.在此基础上给出了单个对象的缓存价值函数,并设计了RAC替换算法.实验表明RAC具有较好的资源适应性和系统吞吐率.     

P2P中一种使用缓存和测量的节点选择模式

针对非结构化P2P中服务节点的选择问题,利用缓存和测量技术,提出了测量次数可变的选择模式(VM)和测量次数不变的选择模式(FM).同时,给出了4种缓存更新策略:随机丢弃,最大延迟丢弃,最早时间丢弃和最少使用丢弃.仿真实验的结果表明,FM优于VM,最大延迟丢弃和最少使用丢弃策略获得了较好的性能,同时,该选择模式对节点加入和离开有较好的鲁棒性,且并发测量可以降低节点选择的时延.