路由器缓存大小设置策略的研究综述

首先介绍了路由器缓存大小设置问题的研究进展,重点讨论了基于TCP协议模型的5种典型的缓存设置方法,通过分析和比较得出基于不同的假设前提的缓存设置方法适应于不同的网络环境;并进一步研究了影响缓存需求的几个主要因素,着重分析了各种TCP协议、队列管理机制与路由器缓存大小设置策略的相互影响,并指出在"-3前复杂时变网络特征下需要采用根据网络状态动态调节阅值的缓存设置方法.     

小缓存对网络实时业务的影响

论述基于TCP协议模型的各种骨干路由器缓存容量研究成果。针对网络实时视频业务的发展,研究小缓存对网络实时视频流的影响,总结小缓存存在的不足,利用网络仿真工具NS2分析小缓存对网络性能的影响。实验结果表明,当链路利用率较低时,小缓存能够满足实时业务的性能要求。     

核心路由器中缓存容量问题的研究

目前核心路由器在网络拥塞时用以存储数据的缓存容量巨大,这虽然保持了极高的链路利用率,但使核心路由器的结构更为复杂。斯坦福法则认为路由器中缓存容量可减少为拇指法则的1%。该文指出斯坦福法则会引起较高的丢包率,提出在降低一定链路利用率的情况下,较小的缓存容量也可以获得较低的丢包率和较高的吞吐率。     

多维交换网络中的一种流量控制机制

多维交换网络是大容量、可扩展分组交换机/路由器设计中的关键技术之一.本文给出了基于这种网络拓扑、采用虫孔路由、结合虚通道技术时的一种有效的流量控制机制,它可在这种网络中同时有效地实现分组级和微片级的流量控制,提高网络交换性能.文中对节点缓存长度与网络性能进行了分析和仿真,描述了有限缓存资源下单节点和全网的流量控制过程,给出了网络中发生阻塞的条件.     

ICN中基于节点权重的缓存大小分配方案研究

网内缓存功能是信息中心网络ICN（Information-Centric Networking）最重要的特性之一,大大减小了信息请求的响应时间和网内流量。合理地分配每个路由器的缓存空间大小,对网络性能有较大影响,也可以节约网络成本。为了使路由器的缓存大小配置合理,首先综合考虑路由器的度数权重、紧密度、网络的中心度、请求影响度等度量指标,定义了一个新的度量指标,称为节点权重;然后,提出一种基于节点权重的缓存大小分配方案,将网络所需的容量按比例分配给路由器。仿真结果表明,与均匀分配相比,路由器的缓存空间利用率至少提升了8%,命中率至少提高了6%;与基于请求影响度的分配方案相比,路由器的缓存空间利用率至少提升3%,命中率至少提高了3%。     

基于可重构路由器上缓存的流媒体协作分发策略刍议

基于可重构路由器上缓存的流媒体协作分发目前出现了一种新的策略,用以加速流媒体。主要就是利用网络存储通过多个路由器节点进行合作,对于一些热点的视频数据实现了缓存,能够为范围较小的用户服务,大大降低了流媒体服务器的带宽和性能要求,降低了网络传输流量与用户的相应延迟时间。本文利用新型的原型系统对基于可重构路由器上缓存的流媒体协作分发策略进行性能评价,事实证明这种方法在各个方面都有所提高。     

基于周期性对数流量模型的路由器性能分析

网络流量模型是网络规划设计和性能分析的基础问题，该文在分析实际网络测试所得数据的基础上提出一个周期性对数突发流量模型，不同于传统的基于概率统计模型，该模型主要考虑网络网络流量的周期性和突发特性，在测试流量数据的基础上通过函数拟合的方法解流量突发上界的时间特性，应用该模型给出了在已知模型参数的条件下，网络互连的基本单元－－路由器输入，输出接口的传输能力，缓存大小以及交换单元处理能力等必须满足的性能指标。     

路由器缓存管理算法之比较研究*

缓存管理是高性能路由器需要解决的技术难题之一,一个好的缓存管理算法可提高路由器的缓存资源利用率并降低分组丢失率.简要介绍了路由器中缓存管理的发展过程,列举了缓存管理一些最主流的算法,并对它们的性质、优缺点作了较为深刻的比较研究.最后利用试验仿真对四种缓存管理算法进行了缓存利用率和分组丢失率方面的评价,并对缓存管理算法的发展作了展望.     

基于分数Alpha模型的缓存计算方法

缓存大小计算是高性能路由器设计中一个必不可少的内容，常规缓存计算方法是基于Poisson通信量模型得到的，不符合网络通信量的实际特征，在使用中存在丢包率较高的问题。已提出的分数Alpha通信量模型能体现通信量的自相似性和非高斯特征，用于缓存溢出概率计算，能得到比其它网络模型更好的结果。基于该模型得到了一个缓存计算方法，在高速路由器转发引擎的缓存设计应用中得到了满意的结果，与常规方法相比更体现了实际通信量的变化规律，计算结果更准确。     

一种缩减内容中心网络的AS域间流量的机制

内容中心网络(content-centric networking,CCN)的一个主要特点就是利用网络内部存储提高网络性能和缩减网络冗余流量.但由于当前路由器存储技术的限制,使得CCN路由器的内容存储器(CS)空间大小十分有限,这造成了路由器缓存的命中率很低,导致CCN的网络性能优势和冗余流量缩减的作用无法得到体现.为了有效地解决这个问题,提出了一种新颖的存储系统AS-BOOST,该系统可以避免当前存储技术的限制,使得CCN的优势得到充分发挥,同时可以防止针对CCN的缓存污染攻击.通过仿真来验证AS-BOOST的有效性,仿真结果表明,只要付出少量的存储开销,AS-BOOST就可以有效地提高用户请求的命中率和明显降低AS域间流量,并且不易受到缓存污染攻击.     

Communication storage optimization for static dataflow with access patterns under periodic scheduling and throughput constraint

We address a recently introduced static dataflow model: the Static Dataflow with Access Patterns (SDF-AP) model. For this model we present (1) a generalization of an existing regular periodic scheduling scheme to regular 1-periodic scheduling for flexibility to achieve a smaller schedule period and additional room for optimization on communication storage; (2) a method based on Integer Linear Programming (ILP) to minimize communication buffers under periodic scheduling and user-specified throughput constraints. Experimental results on a set of test cases show that buffer sizes using this approach can be reduced dramatically when compared to the traditional SDF models. The optimal sizing result may serve as an important criterion to evaluate and fine-tune any heuristics-based buffer sizing approach for the SDF-AP model of computation.     

Decomposition/aggregation-based dynamic programming optimization of partially homogeneous unreliable transfer lines

The general problem of buffers sizing for mean work in process/inventory minimization in a particular class of single part unreliable manufacturing flow lines, subjected to a constant rate of demand for finished parts, is analyzed. Two variants of the problem are considered: buffers sizing for average work in process minimization when there is a fixed requirement on parts availability in the buffer next to last; minimization of an aggregate measure of average work in process and demand backlog when the complete flow line is considered. A fluid model of part production is employed. The production control policies of interest are suboptimal, strictly decentralized, and are unambiguously parameterized by the size of buffer levels. Optimization of policy parameters is based on the analysis of the structural properties of an associated dynamic program. The latter is built around an approximate, flow line decomposition based, buffer levels dependent theoretical expression of the policy performance measure. The nature of the related flow line approximations is discussed and numerical results of the dynamic programming procedure are reported. Scalability of the computations is demonstrated. The numerical results suggest that when parameters are optimal, both a form of flow line balancing and a just in time internal production principle, are in place.     

Sizing and grading for wearable products

Sizing and grading are widely used to create products to fit selected populations. Currently, the sizing and grading rules are derived from anthropometric measures; however past researches have indicated that it is not very accurate. This study proposes a new technique to use principal component analysis (PCA) on 3D surface points for sizing and grading wearable products. The accuracy of the proposed method is illustrated by developing a sizing and grading rule for the feet. After developing a model using the feet data of 60 participants and validating using the feet data of 10 different participants, results showed that sizing and grading using PCA is more accurate than traditional techniques. Compared with traditional foot sizing, PCA based sizing and grading showed an improvement of about 25% in accuracy. In addition, results also indicated that the grading rule derived from PCA loading was better than the proportional grading. This research provides a new direction to consider when developing the sizing and grading rules. It can be extended to calculate the number of sizes and the size increment for various wearable products.     

一种基于不确定规则的数据时效性判定方法

数据过时是影响数据质量的重要因素,因此判定数据时效性对于提高数据质量至关重要.当前判定数据时效性的方法可分为两类:基于时间戳的方法和基于规则的方法.基于时间戳的方法要求精确完整的时间戳,但这样的时间戳在很多应用中不存在.基于规则的方法不要求时间戳,但现有方法均依赖于冗余元组,且不能对数据时效性做出定量判定.同时,这些方法均基于确定规则,无法表达不确定的领域知识.针对上述问题,提出不确定时效规则及相应的数据时效性模型.基于该模型,进一步给出了两个可定量地判定数据时效性的算法.同时,还给出了时效规则的学习算法.真实数据上的实验结果验证了算法的有效性.     

Analysis of dynamic control strategies of an FMS under different scenarios

This paper presents a simulation study aimed at evaluating the performances of a flexible manufacturing system (FMS) in terms of makespan, average flow time, average delay time at local buffers and average machine utilization, subject to different control strategies which include routing flexibilities and dispatching rules. The routing strategies under evaluation are ‘no alternative routings’; ‘alternative routings dynamic’; and ‘alternative routings planned’. Above routing strategies are combined with seven dispatching rules, and studied in different production volume which varies from 50 to 500 parts. In addition, impacts of both infinite and finite local buffer capacities are analyzed. Since an FMS usually deals with a variety of products, effects of changing the part mix ratio are also discussed. Finally, machine failure is also introduced in this research to study the effects of machine reliability on the system. Simulation results indicate that the ‘alternative routings planned’ strategy outperforms other routing strategies if the local buffer size is infinity. However, there is no particular dispatching rule that performs well in all buffer size settings but infinity buffer size is not the best choice with respect to the four performance measures. In addition, the four performance measures, except machine utilization, under different control strategies seem quite insensitive to the variation in part mix ratios.     

TCP流竞争拥塞及拥塞链路的缓存需求研究

针对流竞争拥塞,提出了一种拥塞分析模型FCCM(flow-competing congestion model),给出了TCP竞争流在拥塞链路上的分布特性,推导了流竞争拥塞发生的条件,进而分析了在流竞争拥塞发生时,路由器为维持拥塞链路100%利用率所需的最小缓存.分析结果表明,当流数目不确定时,应对流竞争拥塞所需的缓存将不大于流数目确定时经典BSCL(buffer sizing for congested interact links)方案中的最小缓存需求.     

Discovering Neglected Conditions in Software by Mining Dependence Graphs

Neglected conditions are an important but difficult-to-find class of software defects. This paper presents a novel approach to revealing neglected conditions that integrates static program analysis and advanced data mining techniques to discover implicit conditional rules in a code base and to discover rule violations that indicate neglected conditions. The approach requires the user to indicate minimal constraints on the context of the rules to be sought, rather than specific rule templates. To permit this generality, rules are modeled as graph minors of enhanced procedure dependence graphs (EPDGs), in which control and data dependence edges are augmented by edges representing shared data dependences. A heuristic maximal frequent subgraph mining algorithm is used to extract candidate rules from EPDGs, and a heuristic graph matching algorithm is used to identify rule violations. We also report the results of an empirical study in which the approach was applied to four open source projects (openssl, make, procmail, amaya). These results indicate that the approach is effective and reasonably efficient.     

Associative classification of mammograms using weighted rules

In this paper, we present a novel method for the classification of mammograms using a unique weighted association rule based classifier. Images are preprocessed to reveal regions of interest. Texture components are extracted from segmented parts of the image and discretized for rule discovery. Association rules are derived between various texture components extracted from segments of images and employed for classification based on their intra- and inter-class dependencies. These rules are then employed for the classification of a commonly used mammography dataset, and rigorous experimentation is performed to evaluate the rules’ efficacy under different classification scenarios. The experimental results show that this method works well for such datasets, incurring accuracies as high as 89%, which surpasses the accuracy rates of other rule based classification techniques.