基于Linux的HPCC的设计与实现

集群是目前最广泛被采用的高性能计算机系统解决方案,构建一个HPCC需要多个节点协同安装和配置,本文介绍OSCAR各部分功能,说明用OSCAR套件构建HPCC的过程,最后用实例验证HPCC的成功构建.     

基于虚拟化技术的Rocks高性能集群的建立与应用

集群是目前高性能计算机系统主要的解决方案。随着集群规模的扩大。也出现了不易安装与管理、故障率高、缺乏方便的并行程序开发调试环境等问题。本文在集群安装软件包Rocks的基础上。结合Xen虚拟化技术构建了一个高性能虚拟集群,从而简化了集群的组建与管理,并提高了系统可靠性与容错性能及并行程序开发效率。     

集群式高性能计算系统研究

研究了集群的系统结构和主要优势,以及集群式高性能计算系统的诞生;分析了集群式高性能计算系统的架构和构建方式,集群构建包括网络部署、存储系统、计算节点、管理节点、登录节点等部分。在此基础上构建了基于Linux的集群式高性能计算系统。     

基于Rocks的高性能集群平台搭建与应用

高性能计算是实现大规模并行计算的有效途径,本文论述了如何构建一个高性能集群,介绍了集群安装工具--Rocks,分析Rocks安装机制、特点以及Rocks如何管理集群,利用Rocks搭建了一个基于Linux的高性能集群平台,并在该平台上实现了量子化学软件Gaussian的应用,进一步测试了Rocks系统的性能。     

广域网环境下应用系统远程部署的研究及实现

在集群的众多接点上人工部署和维护应用系统是件繁重的工作。应用远程部署系统利用广域互联网络，通过机群管理中心的主监控台来自动完成对集群中的任一节点安装操作系统、安装升级或删除应用软件等任务。该系统可以快速地同时对多个节点安装Linux操作系统，支持安装的Linux操作系统有多个版本，具有良好可扩展性，能帮助用户显著地减少管理维护费用和提高整个系统的可用性。     

基于MPICH2的高性能计算集群系统研究

目前在高等学校和科研机构中对于高性能计算的需求很大,而商业的超级计算机性能虽高但价格昂贵,同时这些单位又都拥有大量普通的PC机和网络设备.为了利用现有硬件资源获取高性能计算能力,文中研究了在PC机和Linux环境下构建基于MPICH2的高性能计算集群系统的方法,搭建了一个拥有16个节点的系统并利用高性能Linpack基准测试方法进行了性能测试.测试结果表明,这种构建高性能计算集群系统的方法切实可行,是低成本获取高性能计算能力的良好途径.     

一种集群服务器配置和管理工具

文中设计和实现了一个集群服务器的配置和管理的工具，可以利用它来创建一个单系统映像集群服务器系统，并对集群系统的全局状态信息，后台各个单独节点的状态信息进行监控，对后台节点的运行状态进行控制。     

一个可靠的集群簇/网格监控系统

以开放源代码的Ganglia为基础设计和实现了一个具有更高可靠性的集群簇／网格监控系统，该系统能够抵御汇集节点和与之相连的线路的失效。并对可靠性的设计、定量分析与实现，性能的理论分析与实际测试作了论述。理论分析和实验都表明，系统的实现在几乎没有牺牲性能的情况下，取得了更高的可靠性。     

提高集群系统I／O节点吞吐能力的方法研究

I/O节点是Linux高性能集群系统中必不可少的组成部分,也常常成为高性能集群系统的瓶颈。文章介绍了通过多网卡捆绑来提高I/O节点吞吐能力的方法及其实现过程。实践证明,这是一种非常有效和低成本的解决方案。     

Linux高性能计算集群的设计与实现

计算机和网络硬件设备逐步实现商品化和标准化,PC机或工作站的性能越来越高而价格越来越便宜,同时开源Linux微内核及集群工具中间件技术也日趋成熟稳定,高性能计算集群逐渐发展起来,并成为主流的高性能计算平台。高性能计算集群逐渐替代专用、昂贵的超级计算机对大规模并行应用构建原型、调试和运行。基于PCs或工作站的高性能计算快速部署及其可靠性和可管理性研究,对高性能计算集群在科学研究和工程计算等领域的应用,促进高性能计算技术的应用方面具有深远的意义。本文以OSCAR集群为实例,部署一个五结点的集群环境并运行简单的并行测试例子。     

基于异构机群的高速网络入侵检测系统

结合异构机群系统,提出一种基于双向驱动的分流算法,将高速数据流分为多个子数据流,把子数据流交由异构机群系统中最合适的节点处理,实现基于异构机群的高速网络入侵检测系统。实验结果表明,该系统保证了某时间段内具有相同源或目的地址的所有数据包发向同一个后端IDS引擎进行检测,能在高速网络环境下保持高检测率,并有效解决负载均衡问题。     

Visual control of two aerial micro-robots by insect-based autopilots

In the framework of our research on biologically inspired microrobotics, we have developed two novel Automatic Flight Control Systems (AFCs): OCTAVE ( Optical flow Control sysTem for Aerospace VEhicles) and OSCAR ( Optical Scanning sensor for the Control of Autonomous Robots), both based on insects' visuomotor control systems. OCTAVE confers upon a tethered aerial robot (the OCTAVE robot) the ability to perform terrain following. OSCAR gives a tethered aerial robot (the OSCAR robot) the ability to fixate and track a contrasting target with a high level of accuracy. Both OCTAVE and OSCAR robots are based on optical velocity sensors, the principle of which is based on the results of previous electrophysiological studies on the fly's Elementary Motion Detectors (EMDs) performed at our laboratory. Both processing systems described are light enough to be mounted on-board Micro-Air Vehicles (MAVs) with an avionic payload small enough to be expressed in grams rather than kilograms.     

Application placement in computer clustering in software as a service (SaaS) networks

One major service provided by cloud computing is Software as a Service (SaaS). As competition in the SaaS market intensifies, it becomes imperative for a SaaS provider to design and configure its computing system properly. This paper studies the application placement problem encountered in computer clustering in SaaS networks. This problem involves deciding which software applications to install on each computer cluster of the provider and how to assign customers to the clusters in order to minimize total cost. Given the complexity of the problem, we propose two algorithms to solve it. The first one is a probabilistic greedy algorithm which includes randomization and perturbation features to avoid getting trapped in a local optimum. The second algorithm is based on a reformulation of the problem where each cluster is to be assigned an application configuration from a properly generated subset of configurations. We conducted an extensive computational study using large data sets with up to 300 customers and 50 applications. The results show that both algorithms outperform a standard branch-and-bound procedure for problem instances with large sizes. The probabilistic greedy algorithm is shown to be the most efficient in solving the problem.     

基于网格节点的分布计算策略分析

本文介绍了网格技术的现状和近期发展,讨论了集群服务器作为网格节点的重要组成部分在网格计算中的重要作用,重点给出了计算包在到达网格节点时的一种数学模型,并利用这种数学模型对网格节点的分布计算策略进行了分析。这对充分利用集群服务器计算资源、提高网格节点的服务能力有着重要意义。     

‘Best’ for whom?: the tension between ‘best practice’ ERP packages and diverse epistemic cultures in a university context

The idea that so-called ‘best’ business practices can be transferred to organizations when they purchase enterprise resource planning (ERP) software packages is a major selling point of these packages. Yet recent research has illustrated a gap between the espoused theory of a best practice solution and the theory-in-use experienced by those who install software with such a design. As researchers begin to examine the difficult process by which organizations recast the best practices model handed down to them by consultancies and software vendors in an effort to make the software ‘work for them’ in practice, it is equally important that we begin to understand the reasons that such a gap exists. To this end, we analyze the strategic partnership between a multinational software vendor and a university who together designed a ‘best practice’ ERP package for the higher education industry. Through the theoretical lens of ‘epistemic cultures’ we argue that in organizational contexts made up of more than one epistemic culture, the use of a best practice model will be problematic because, by definition, the model mandates one epistemological position through the software design. This is counter to a university's loosely coupled organizational form.     

基于信任的簇技术在MANET入侵检测中的应用

MANET（a Mobile Ad hoe Network）工作在没有固定设施的情况下，所以很具竞争优势。然而这种优势伴随着安全方面的巨大挑战，尤其对串通勒索攻击不具免疫能力。提出一种建立在簇和信任概念上的新的入侵检测机制来解决这个问题，这里的信任概念能弥补已经存在的能保证MANET安全的那些预防机制的缺陷。实验结果证明：通过簇内的信任机制，由勒索攻击者所造成的伪造指控和伪造警告问题可以被去除。     

一种支持SDAI的数据库模式定义语言

目前，许多应用领域，如ＣＡＤ、ＣＡＳＥ、ＭｕｌｔｉＭｅｄｉａ、ＭＩＳ等对数据库有了新的要求，这就使得面向对象的数据库成为国内外数据处理领域中非常具有诱惑力的一个研究方向。     

A comparative study of the reliability of nine statistical software packages

The reliabilities of nine software packages commonly used in performing statistical analysis are assessed and compared. The (American) National Institute of Standards and Technology (NIST) data sets are used to evaluate the performance of these software packages with regard to univariate summary statistics, one-way ANOVA, linear regression, and nonlinear regression. Previous research has examined various versions of these software packages using the NIST data sets, but typically with fewer software packages than used in this study. This study provides insight into a relative comparison of a wide variety of software packages including two free statistical software packages, basic and advanced statistical software packages, and the popular Excel package. Substantive improvements from previous software reliability assessments are noted. Plots of principal components of a measure of the correct number of significant digits reveal how these packages tend to cluster for ANOVA and nonlinear regression.     

工程数据库管理系统OSCAR的页面服务器

为了适应工程应用的特点，工程数据库管理系统OSCAR采用了页面服务器体系结构．OSCAR的页面服务器集中实现了数据库管理系统核心的事务管理，即并发控制和恢复功能．它向客户端提供支持多用户共享的、可靠的数据页面服务，客户端模块在此基础上构建一个功能完备的工程数据库管理系统．介绍了设计与实现OSCAR页面服务器过程中碰到的问题及其解决方法，包括OSCAR页面服务器的功能在物理层、缓冲层和事务层的分配，客户／服务器环境中ARIES恢复策略的改进以及先写日志文件原则的实现方案。