【摘 要】
：
Communication systems utilize the Distributed Hash Table (DHT) approach to build the network infrastructure for advantages of even distribution of workload,high scalability and cost-effectiveness.Alth
【机 构】
：
State Key Laboratory of Networking & Switching Technology
论文部分内容阅读
Communication systems utilize the Distributed Hash Table (DHT) approach to build the network infrastructure for advantages of even distribution of workload,high scalability and cost-effectiveness.Although DHT is undoubtedly applicative in such architectures,some practical distinctions still should be considered to meet the performance requirements of communication infrastructures.This paper focuses in two features of the distributed communication system,socalled the real-time response and geographic partition,proposes a hierarchical DHT lookup service named Comb.Comb’s overlay is organized as a two-layered architecture,workload is distributed evenly among nodes and most queries can be routed in no more than two hops.Comb performs effectively with low bandwidth consumption and satisfactory fault tolerance even in a continuously changing environment.Both theoretical analyses and experimental results demonstrate that the two-layered architecture of Comb is feasible and efficient.Comb improves the performances on routing delay and lookup failure rates with high scalability and availability.
其他文献
网络入侵的早期特征是影响网络入侵早期检测效果的关键。针对网络入侵早期特征选择问题,提出一种结合频率筛选的遗传算法,该算法以SOM神经网络作为评价模型,通过多次运行遗传算法改善其优化结果的稳定性,根据对最优解中特征出现的频率进一步筛选,得到一组优化的早期特征。对入侵早期特征集进行特征选择实验,将39维早期特征优化至29维。实验结果表明,使用优化特征组合不仅有效缩减了入侵检测建模时间,而且使入侵检测系
在无线传感器网络路由协议中使用分簇技术可以减少节点能量消耗,延长网络生命周期。簇头的选择是分簇路由最重要的部分之一,本文基于LEACH提出一种改进的簇头选择方法LEACH-EP,它的核心是在近K轮工作周期里保持簇的结构不变,除第1轮按照标准的LEACH选择簇头节点外,余下的K-1轮基于能量预测选择簇头节点。该方法减少了簇结构的更新次数,推迟了节点开始死亡的时间。OMNET++模拟结果显示,LEAC
Long-term prediction is a key problem in real-time video traffic applications.Most of real-time video traffic belong to VBR traffic and has specific properties such as time variation,non-linearity and
传统的网络入侵检测方法利用已知类型的攻击样本以离线的方式训练入侵检测模型,虽然对已知攻击类型具有较高的检测率,但是不能识别出网络上新出现的攻击类型.本文提出了一种基于增量式GHSOM神经网络模型的入侵检测方法,在不破坏已学习过的知识的同时,对在线检测过程中新出现的攻击类型进行增量式学习,实现对入侵检测模型的动态扩展.在线入侵检测实验结果表明增量式GHSOM入侵检测方法具有动态自适应性,而且对于网络
Objective and accurate assessment of each node influence is a vital issue to research social networks.Many algorithms have been developed,but most of them use of single metric,which is incomplete and
This paper considers sub-channel and power allocation based on genetic algorithms to maximize the overall system capacity using proportional rate constraints in multiuser orthogonal frequency division
In DTNs,routing protocols use “store-carry-forward” approaches to complete the communication process.As traditional “Connectivity” concept from the Internet ignores the possibility of opportunistic co
Resource reservation is a widely used mechanism in distributed systems and high-performance networks,and the optimization of its performance has been greatly concerned.Data structure is used to store
In virtualized and dynamical cloud computing environment,all resources can be virtualized and provided as IT services which can be accessed through internet in a pervasive way.One can create new value
The P2P (peer-to-peer) has been widely used in file sharing,online chatting,peer computing,etc.Network traffic generated by P2P applications makes up of a large portion of the overall Internet traffic