计算系统虚拟化平台的研究及实现

虚拟化技术是高性能计算系统规模化的关键技术。高能所计算资源虚拟实验床采用 OpenStack 云平台搭建环境。本文讨论了实现虚拟计算资源与计算系统相互融合的三个关键因素：网络架构设计、环境匹配和系统总体规划。本文首先讨论了虚拟网络架构。虚拟化平台通过部署 neutron 组件、OVS以及 802.1Q 协议来实现虚拟网络和物理网络的二层直连,通过配置物理交换机实现三层转发,避免了数据经过 OpenStack 网络节点转发的瓶颈。其次,虚拟计算资源要融入计算系统,需要与计算系统的各个组件进行信息的动态同步,以满足域名分配、自动化配置以及监视等系统的需要。文章介绍了自主开发的 NETDB 组件,该组件负责实现包括虚拟机与域名系统 (DNS)、自动化安装和管理系统 (puppet) 以及监视系统的信息动态同步等功能;最后,在系统总体规划中,文章讨论了包括统一认证、共享存储、自动化部署、规模扩展和镜像等内容。     

主动互联的关键技术

主动互联是下一代网络技术。与传统网络相比，主动网络灵活、可扩展并能动态升级。但是，在主动网络的实现中有诸如安全性、性能、互操作和合适的操作系统等技术上的挑战。该文介绍了主动网络并讨论了这些技术上的挑战，提出了一些解决方法。     

Topological integration of BIM and geospatial water utility networks across the building envelope

Utility networks comprise a fundamental part of our complex urban systems and the integration of digital representations of these networks across multiple spatial scales can be used to help address priority challenges. Deteriorating water utility infrastructure and low routing redundancy result in network fragility and thus supply outages when assets fail. Water distribution network configurations can be optimised for higher resilience but digital representations of the networks used for simulations and analyses are not integrated with the finer scale networks inside buildings. This integration is hindered by differences in conceptualisation and semantics employed by the relevant data standards. We suggest that the geospatial and geometric data contained in Building Information Modelling (BIM) and water distribution network (WDN) models can be used for their integration; and that this supports the use cases of optimising dynamic network partitioning, reducing the risk of underground utility strikes and planning for future network configurations with higher topological redundancy. In this study, we develop and demonstrate the application of a weight-based spatial algorithm for inferring water network connections between urban-scale WDNs and BIM models, showing that spatial data can be used in the absence of complete or consistent semantic representations. We suggest that the method has potential for transferability to infrastructure for other utility resources (such as waste water, electricity and gas) and make recommendations such as standardising the representation of connection points between disjoint utility network models and extending the normal practical spatial remit of BIM MEP modelling to encompass the space between buildings and WDNs.     

Recent Advances in Mobile Grid and Cloud Computing

Grid and cloud computing systems have been extensively used to solve large and complex problems in science and engineering fields. These systems include powerful computing resources that are connected through high-speed networks. Due to the recent advances in mobile computing and networking technologies, it has become feasible to integrate various mobile devices, such as robots, aerial vehicles, sensors, and smart phones, with grid and cloud computing systems. This integration enables the design and development of the next generation of applications by sharing of resources in mobile environments and introduces several challenges due to a dynamic and unpredictable network. This paper discusses applications, research challenges involved in the design and development of mobile grid and cloud computing systems, and recent advances in the field.     

A framework for designing ATM management systems by way of abstract information models and distributed object architectures

ATM network management has not yet reached the level of versatility and comprehensiveness exhibited by other aspects of this networking technology. Since the currently dominant general standards for broadband network management, primarily targeted towards large carrier networks, are too complex and intricate for smaller environments, e.g. LANs or corporate networks, ATM equipment manufacturers, in their effort to provide systems that exploit the power of ATM and yet remain practical and simple, usually resort to devising ad hoc proprietary extensions of simpler management frameworks, originally developed for other networking technologies. Such incompatible extensions, however, remain useless in the common case where the network is heterogeneous. The notion of abstract information modeling may be employed for improving on this situation.Building on this concept, the paper proposes a framework for developing ATM management systems intended for heterogeneous small- to medium-size networks. The general guidelines are illustrated through discussing a specific compliant management application intended for the remote monitoring of ATM network platforms and developed in the framework of a European research project. The application features a WWW interface and, as such, provides an example of the compatibility of information abstraction with the Web-based management techniques and of the benefits arising from the combination of the two notions.Lastly, and as the concept of abstracting information is inherent to recent frameworks for designing and programming distributed object systems, the paper explores this relation by discussing an alternative design of the monitoring application as a distributed object system.     

Activating networks: a progress report

Active networks-networks you can add programs to or customize to particular applications-are probably most familiar as Web proxy caches and firewalls. In their more sophisticated form, however, they have become customized infrastructures that let designers program control planes-the control software and network hardware used to manipulate the transport system's behavior. Programming mechanisms such as open signaling are becoming more widespread, for example. But research in active networks has taken a step beyond even these sophisticated infrastructures. Working systems now let designers modify packet-switching infrastructures on the fly using either a switch-like model, which mixes active packets with other packets, or in a capsule model, which regards all packets as programs. In surveying working systems and experimental results, the authors have found that first-generation systems have opted to use modern programming language technologies such as Caml and Java to provide usability and safety, and cryptographic techniques to provide security. The systems differ in the degree of flexibility and performance they offer. Many differences stem from the use of a particular security model. The paper discusses a general architecture for active networks     

Type-1 and type-2 fuzzy inference systems as integration methods in modular neural networks for multimodal biometry and its optimization with genetic algorithms

We describe in this paper a comparative study between fuzzy inference systems as methods of integration in modular neural networks for multimodal biometry. These methods of integration are based on techniques of type-1 fuzzy logic and type-2 fuzzy logic. Also, the fuzzy systems are optimized with simple genetic algorithms with the goal of having optimized versions of both types of fuzzy systems. First, we considered the use of type-1 fuzzy logic and later the approach with type-2 fuzzy logic. The fuzzy systems were developed using genetic algorithms to handle fuzzy inference systems with different membership functions, like the triangular, trapezoidal and Gaussian; since these algorithms can generate fuzzy systems automatically. Then the response integration of the modular neural network was tested with the optimized fuzzy systems of integration. The comparative study of the type-1 and type-2 fuzzy inference systems was made to observe the behavior of the two different integration methods for modular neural networks for multimodal biometry.     

Secure SCADA framework for the protection of energy control systems

Energy distribution systems are becoming increasingly widespread in today's society. One of the elements that are used to monitor and control these systems are SCADA (Supervisory Control and Data Acquisition) systems. In particular, these control systems and their complexities, together with the emerging use of the Internet and wireless technologies, bring new challenges that must be carefully considered. Examples of such challenges are the particular benefits of the integration of those new technologies, and also the effects they may have on the overall SCADA security. The main task of this paper is to provide a framework that shows how the integration of different state‐of‐the‐art technologies in an energy control system, such as wireless sensor networks, mobile ad hoc networks, and the Internet, can bring some interesting benefits, such as status management and anomaly prevention, while maintaining the security of the whole system. Copyright © 2010 John Wiley & Sons, Ltd.     

Real‐Time Major Events Monitoring and Alert System through Social Networks

The volume of information generated by social and cellular networks has significantly increased in recent years. Automated collection of these data and its rapid analyses allow for better and faster detection of major (in terms of National impact) ‘real life’ events. This study uses data obtained from social networks such as Twitter and Google+. It proposes a mechanism for detecting major events and a system to alert on their manifestation. The article describes the considerations and needed algorithms required to develop and establish such a system. The methodology presented here is based on linking major events that occurred in Israel during the years 2011–2014, with information extracted from social networks. Results indicate that alerts were received shortly after the event occurred for most of major events. Such are large fires, earthquakes and terror attacks. However, attempts to achieve alerts for ‘local’ secondary events failed. This as their impact on the social network is low. Furthermore, it was found that the volume of false alerts depends on the type of domain and keywords.     

Survivability of multiagent-based supply networks: a topological perspect

Supply chains involve complex Webs of interactions among suppliers, manufacturers, distributors, third-party logistics providers, retailers, and customers. Although fairly simple business processes govern these individual entities, real-time capabilities and global Internet connectivity make today's supply chains complex. Survivability is a critical factor in supply network design. Specifically, supply networks in dynamic environments, such as military supply chains during wartime, must be designed more for survivability than for cost effectiveness. We present a methodology for building survivable large-scale supply network topologies that can extend to other large-scale MASs. To create survivable-and hence dependable-multiagent systems, we must also consider the interplay between network topology and node functionalities.     

Modeling complex systems with adaptive networks

Adaptive networks are a novel class of dynamical networks whose topologies and states coevolve. Many real-world complex systems can be modeled as adaptive networks, including social networks, transportation networks, neural networks and biological networks. In this paper, we introduce fundamental concepts and unique properties of adaptive networks through a brief, non-comprehensive review of recent literature on mathematical/computational modeling and analysis of such networks. We also report our recent work on several applications of computational adaptive network modeling and analysis to real-world problems, including temporal development of search and rescue operational networks, automated rule discovery from empirical network evolution data, and cultural integration in corporate merger.     

Controlling across complex networks - Emerging links between networks and control

This paper discusses the interplay between networks and control systems. As we gain more understanding about the structure and dynamics of physical networks, their effects on the performance of closed-loop control systems, as well as the ability to control such networks, provide fertile areas of research. The paper reviews such research with special emphasis on the connectivity and delays in the information transfer across networks.     

A u-IT collaboration evaluation model for value networks

The paradigm of competition between companies has been changed from the competition between individual businesses to the one between the supply chain management. To gain a comparative advantage in this supply chain management, IT collaboration, which has an important strategic location for cooperation between manufacturers, sectors, systems, applications, and processes and supports them, has begun to emerge as an important key. To date, however, most of the studies have searched factors constructing collaborative networks between companies, and there are few studies on performance analysis for activating collaborative networks between companies. Accordingly, this study would like to conduct performance analysis on collaborative networks between companies for measuring performance of constructing concrete collaborative networks and activating improved collaborative network systems from an industrial viewpoint. In detail, it would like to design a model for determinants to make construct collaborative networks and to analyze the utilization level and performance of collaborative networks between companies. In addition, it conducts correlation analysis on determinants of collaborative networks to improve performance of these collaborative networks.     

Trust Management in Distributed Systems

Distributed systems such as the Internet, peer-to-peer networks, and mobile ad hoc networks involve numerous entities, many of which haven't previously interacted. Trust management can help minimize risk and ensure the network activity of benign entities in distributed systems. This article presents different trust models. It also discusses trust management in P2P system     

Linking market uncertainty to VoIP service architectures

The flexibility of packet-switching technologies, coupled with the economic benefits of sending voice over IP networks, is accelerating the convergence of data and voice. It remains unclear, however, which voice-over-IP (VoIP) architecture will best meet user needs, which services will be most successful, and where in the network these services should reside. The question of where to place services in IP telephone networks depends on many variables, including the end systems' capabilities, the amount of interaction the service requires with the end user, and the network infrastructure's architecture. The limited ability of dumb end devices, such as the telephones in the public switched telephone network (PSTN), force service providers to implement services within the network. Smart end systems allow providers more choice in where to locate services. This article describes the main architectural options for VOW and discusses their associated economic effects. By focusing on the relationship between architecture and market uncertainty, the author demonstrates how high uncertainty enhances the value of architectures that allow distributed network services.     

Memory neuron networks for identification and control of dynamicalsystems

This paper discusses memory neuron networks as models for identification and adaptive control of nonlinear dynamical systems. These are a class of recurrent networks obtained by adding trainable temporal elements to feedforward networks that makes the output history-sensitive. By virtue of this capability, these networks can identify dynamical systems without having to be explicitly fed with past inputs and outputs. Thus, they can identify systems whose order is unknown or systems with unknown delay. It is argued that for satisfactory modeling of dynamical systems, neural networks should be endowed with such internal memory. The paper presents a preliminary analysis of the learning algorithm, providing theoretical justification for the identification method. Methods for adaptive control of nonlinear systems using these networks are presented. Through extensive simulations, these models are shown to be effective both for identification and model reference adaptive control of nonlinear systems.     

An empirical study of the structure of relevant keywords in a search engine using the minimum spanning tree

This paper provides a comprehensive study of the structure of relevant keywords in a search engine using the minimum spanning tree (MST) approach. In the process of constructing MST’s, we introduce a novel metric to measure a distance between keywords by applying an integration of the Pearson correlation and the query-based cosine similarity. From this work, we made several meaningful observations about the networks of relevant keywords. First, keyword networks in a search engine exhibit the small-world effect and the scale-free property. Second, only a few among relevant keywords in the order of popularity are positively correlated and there is no significantly positive or negative relationship for the rest of relevant keywords. Third, the degree of searching activity for relevant keywords varies depending on whether they are branded keywords or non-branded keywords as well as the characteristics of product categories. Fourth, the mean correlation coefficient for keyword impressions during slow season increases. Finally, both k_max and the betweenness centrality for high-involvement products are higher than those for low-involvement products.     

On the sample complexity of learning for networks of spiking neurons with nonlinear synaptic interactions

We study networks of spiking neurons that use the timing of pulses to encode information. Nonlinear interactions model the spatial groupings of synapses on the neural dendrites and describe the computations performed at local branches. Within a theoretical framework of learning we analyze the question of how many training examples these networks must receive to be able to generalize well. Bounds for this sample complexity of learning can be obtained in terms of a combinatorial parameter known as the pseudodimension. This dimension characterizes the computational richness of a neural network and is given in terms of the number of network parameters. Two types of feedforward architectures are considered: constant-depth networks and networks of unconstrained depth. We derive asymptotically tight bounds for each of these network types. Constant depth networks are shown to have an almost linear pseudodimension, whereas the pseudodimension of general networks is quadratic. Networks of spiking neurons that use temporal coding are becoming increasingly more important in practical tasks such as computer vision, speech recognition, and motor control. The question of how well these networks generalize from a given set of training examples is a central issue for their successful application as adaptive systems. The results show that, although coding and computation in these networks is quite different and in many cases more powerful, their generalization capabilities are at least as good as those of traditional neural network models.     

基于神经网络的复杂垃圾信息过滤算法分析

垃圾信息的识别是自然语言处理方面主要的任务之一.传统方法是基于文本特征或词频的方法,其识别准确率主要依赖于特定关键词的出现与否,存在对关键词识别错误或对未出现关键词的垃圾信息文本识别能力较差的问题,提出基于神经网络的方法.首先,利用传统方法针对这一类垃圾信息文本进行识别训练和测试;然后,利用从垃圾短信、广告和垃圾邮件数...     

Metaheuristics for protecting critical components in a service system: A computational study

Deliberate sabotages and terrorist attacks are major threats to the safety of modern societies. These attacks often target at important infrastructures such as energy production and transmission systems, food and water supply networks, telecommunications networks, transportation networks, etc. In such systems, some components are critical as their malfunction may adversely affect the operations of the whole systems. This research examines several models based on the median problem for identifying these components in a service system. In addition to the existing exact solution methods, we propose meta-heuristics to tackle this computationally hard problem. Our hybrid approach combines the strengths of both meta-heuristics and exact solution methods. The experiment shows that the combination of solution methods significantly cut down the computational requirement for finding the critical components in a service network for protection.