网络可靠性安全性设计和容错

由于连入Internet,为用户提供各种信息服务。资源共享和开放是Internet特点,因而Internet的安全机制很松散,而用户对网络信息系统要求有较高的安全性,其内部的许多数据和文件严禁未经授权的访问。因此,设计与开发保证内部各种信息的安全机制是实现该网络顺利运行的关键。     

Fault‐tolerant display image‐data‐manipulation unit for system‐on‐panel

Abstract— Bit‐partitioned and conventional shifts, as well as type transformations of multimedia data, are frequently used for display image‐processing systems. A data manipulation unit with fault‐recovery capability based on redundancies is proposed for system‐on‐panel with low processing technology yield. Utilizing data manipulations that are similar to normal shift operation, a proposed data‐manipulation unit is designed with a few additional paths added to the existing barrel shifter. The design methodologies are verified with FPGA and the performance is evaluated in terms of the advantages.     

On basic properties of fault-tolerant multi-topology routing

Multi-topology routing has recently gained popularity as a simple yet efficient traffic engineering concept. Its basic purpose is to separate different classes of network traffic, which are then transported over disjoint logical topologies. Multi-topology routing is used as a basis for implementation of an IP fast reroute scheme called Multiple Routing Configurations (MRC).MRC has a range of attractive properties, but they do come at a cost. In order to guarantee recovery from any single link or node failure in the network, MRC has to maintain several logical topologies and thus an increased amount of routing information. The number of the logical topologies in MRC need not be large; even simple heuristic algorithms often yield good results in practice. However, why this is the case is not fully understood yet.In this paper, we introduce a theoretical framework for fault-tolerant multi-topology routing (FT-MTR). MRC is a practical implementation of FT-MTR in connectionless IP networks. We use FT-MTR to study how the internal topological structure of the communication network relates to two important problems. The first problem is minimizing the number of logical topologies and thus the routing state in FT-MTR. We show how to use the sets of nodes that separate the topology graph to devise an advanced heuristic for “intelligent” construction of the logical topologies. Finding the separating sets in a topology graph is computationally demanding; we present an algorithm that performs well in tested real network topologies. We evaluate the separation-set based heuristic for the logical topology construction and show that it outperforms the known MRC heuristics.The second problem is the FT-MTR load distribution after a failure. We use the separating sets to devise a novel algorithm for failure load distribution. This algorithm does not require knowledge of the traffic demand matrix, still, our tests indicate that it performs as good as, or better than, known MRC load-distribution algorithms that do require the demand matrix as input.     

Optimal decision rules acquisition in incomplete and Fuzzy Decision Information System

The Incomplete and Fuzzy Decision Information System （IFDIS） with both missing data and fuzzy decisions is rather extensively applications, which is the generalizations of complete information system, incomplete information system and fuzzy decision information system. In this paper, on the base of notion of the tolerance relation and the concept of IFDIS, the rough set model of IFDIS based on tolerance relation is proposed and the properties of the models are discussed, and then the definitions of（upper- or lower-） consistent reduction are suggested and knowledge reduction based on discernibility matrix is presented. The optimal fuzzy decision rules and its acquisition algorithm are proposed. Finally we provide an example to illustrate the validity of the algorithm.     

一种新的容错ATM交换机结构的冗余路径分析

异步传输模式ATM是国际化电信联盟建议用于宽带综合业务数字网（B－ISDN）的传输模式。容错是提高网络可靠性的主要方法。该文讨论了一种新的容错ATM交换机结构。该结构以Banyan交换机结构为基础,通过增加子交换单元和扩展链路来增加冗余路径的数目,该结构可以容错多个故障,并且冗余路径的数目随着交换机级数的增加而指数增长。同现有文献中其它容错ATM交换机对比分析可以看出,在具有相近的交换元素个数的前提下,建议的交换机提供了更高的容错能力。     

增强前馈神经网络对多点断路故障的容错能力

神经网络在发生多点断路故障时,网络中的多个隐层神经元及其相关的连接权同时失效。该文针对陷层神经元可以动态增加的一类前馈神经网络,提出了一种三阶段方法T3。T3先对网络进行一次训练,然后用验证集进行测试以确定网络的故障曲线拐点,在此基础上通过第二次训练自适应地增加冗余隐层神经元。实验表明,T3使用较小的冗余量就可以显著改善网络对多点断路故障的容错性,在网络的容错能力和结构复杂度之间较好地达成了折衷。     

一个多机容错系统MPFTS故障恢复的设计与实现

本介绍我们研制的一个多机容错系统MPFTS的故障恢复技术。首先对MPFTS的结构和工作过程进行了概要介绍。在此基础上,分析了该系统不同类型故障恢复的可能性、恢复时机的选择、恢复所要达到的目标等问题。中详细讨论了利用异常中断返回技术实现系统故障恢复所涉及的数据结构及具体运行过程,并对上电升级重构这种特殊的故障恢复情况进行了分析。     

Probabilistic analysis on mesh network fault tolerance

Mesh networks are among the most important interconnection network topologies for large multicomputer systems. Mesh networks perform poorly in tolerating faults in the view of worst-case analysis. On the other hand, such worst cases occur very rarely in realistic situations. In this paper, we study the fault tolerance of 2-D and 3-D mesh networks under a more realistic model in which each network node has an independent failure probability. We first observe that if the node failure probability is fixed, then the connectivity probability of these mesh networks can be arbitrarily small when the network size is sufficiently large. Thus, it is practically important for multicomputer system manufacture to determine the upper bound for node failure probability when the probability of network connectivity and the network size are given. We develop a novel technique to formally derive lower bounds on the connectivity probability for 2-D and 3-D mesh networks. Our study shows that these mesh networks of practical size can tolerate a large number of faulty nodes thus are reliable enough for multicomputer systems. For example, it is formally proved that as long as the node failure probability is bounded by 0.5%$0.5\%$, a 3-D mesh network of up to a million nodes remains connected with a probability larger than 99%$99\%$.     

线性系统的容错控制

在这篇文章里 ,我们考虑一类线性系统的鲁棒跟踪控制器设计问题 .对执行器故障和控制面故障情况 ,提出了容错的鲁棒控制器的设计方法 .此设计方法能够保证闭环系统是稳定的     

Kosha: A Peer-to-Peer Enhancement for the Network File System

The storage needs of modern scientific applications are growing exponentially, and designing economical storage solutions for such applications – especially in Grid environments – is an important research topic. This work presents Kosha, a system that aims to harvest redundant storage space on cluster nodes and user desktops to provide a reliable, shared file system that acts as a large distributed storage. Kosha utilizes peer-to-peer (p2p) mechanisms to enhance the widely-used Network File System (NFS). P2P storage systems provide location transparency, mobility transparency, load balancing, and file replication – features that are not available in NFS. On the other hand, NFS provides hierarchical file organization, directory listings, and file permissions, which are missing from p2p storage systems. By blending the strengths of NFS and p2p storage systems, Kosha provides a low overhead storage solution. Our experiments show that compared to unmodified NFS, Kosha introduces a 3.3% fixed overhead and 4.5% additional overhead as nodes are increased from two to sixteen. For larger number of nodes, the additional overhead increases slowly. Kosha achieves load balancing in distributed directories, and guarantees or better file availability.*This work was supported in part by an NSF CAREER award (ACI-0238379).Troy A. Johnson was supported by a U.S. Department of Education GAANN doctoral fellowship.