多核计算环境下快速排序并行算法的实现

研究了快速排序算法,并在其基础上提出了基于多核技术的OpenMP并行编程模型的快速排序算法.实验结果表明,该并行算法具有较高的并行加速比和并行效率.     

改进的按位拆分快速排序算法

在讨论目前已有的快速排序算法的基础上,提出一种新的按位拆分快速排序算法,利用Java实现了算法的并行运算。算法分析和实验结果表明,它的算法时间复杂度可达到O(Kn),排序速度明显优于Quick Sort。     

并行排序算法

本文对目前所流行的各种并行排序算法作了系统地综述,是一篇面向研究的综合报告。其中,包括Batcher和Preparata的排序网络;Stone,Seigel,Kung和Nassimi等的不同互连网络的SIMD机器上的双调排序算法以及目前所知道的最快的Preparata的快速并行排序算法。     

NOW系统上的并行快速排序算法

介绍了在NOW系统上的并行快速排序算法的设计与实现，分析了影响算法性能的因素及改进方法，最后给出了该算法对字符串排序的并行效率为49．15％。     

基于流水光总线阵列上的快速可扩展并行排序算法

在Y．Pan提出的基于流水光总线阵列模型(LARPBS)上使用N个处理器对N个元素进行排序在最好情况下以O(logN)时间，最坏情况下以O(N)时间完成的并行排序算法的基础上，提出了一种LARPBS模型上的可扩展的快速并行排序算法，对N个元素进行排序，使用p(1≤P≤N)个处理器在最好情况下以O(NlogN／p)时间，最坏情况下以O(N^2／p)时间完成排序。另外还提出了一种LARPBS模型上改进的快速高效并行排序算法，该算法对N个元素进行排序使用N个处理器在最好情况下以O(log√N)时间、最坏情况下以O(√N)时间完成排序。     

基于Java的按位拆分快速排序并行算法

针对大数据量排序算法优化问题,提出一种基于Java的按位拆分的排序新算法。该排序算法按照位拆分数据,并结合Java的多线程对拆分的数据进行并行处理。数据实验结果表明,对于大数据量排序,该算法性能明显优于快速排序算法,而且算法具有很好的并行效率。     

基于CUDA的并行K-近邻连接算法实现

针对大规模空间数据的K-近邻连接查询问题,设计了一种CUDA编程模型下K-近邻连接算法的并行优化方法。将K-近邻连接算法的并行过程分两个阶段:1)对参与查询的数据集P和Q分别建立R-Tree索引;2)基于R-Tree索引进行KNNJ查询。首先根据结点所在位置划分最小外包框,在CUDA下基于递归网格排序算法创建R-Tree索引。然后在CUDA下基于R-Tree索引进行KNNJ查询,其中涉及并行求距离和并行距离排序两个阶段:求距离阶段利用每一个线程计算任意两点之间的距离,点与点之间距离的求取无依赖并行;排序阶段将快速排序基于CUDA以实现并行化。实验结果表明,随着样本量的不断增大,基于R-Tree索引的并行K-近邻连接算法的优势更加明显,具有高效性和可扩展性。     

改进的并行快速排序

排序是计算机内一类很耗时的问题,快速排序是众多排序方式中相对较省时的算法,但当数据量大时,其耗时还是惊人的。为了解决大数据快速排序耗时的问题,论文在常用的快速排序并行方法的基础上,给出了一种改进方法,使其在时间复杂度上有明显的改进。     

基于图形硬件的双调排序算法优化

介绍一种新的并行排序算法,该算法以双调归并排序为基础,运用图形硬件的并行体系结构和二叉排序树数据结构的优点,用部分并行代替所有阶段的顺序执行,对双调排序算法进行优化.对该算法进行分析,在理论上n个序列在P个流处理器上的排序,最优的时间复杂度为O((nlogn)/p).实验测试结果表明,优化后的算法比其它基于图形硬件的双调归并排序算法所用时间短.     

排序合并Join算法的新结果

Join操作是数据库中最昂贵和最常用的操作.排序合并Join算法是实现Join操作的重要算法,得到了普遍接受并广为应用.在重新研究了排序合并Join算法后发现,同时对两个Join关系进行外排序是不必要的,会带来很大的额外开销.针对这个问题,提出了一种基于单关系外排序的分治Join算法,并在该算法的基础上提出了基于单关系外排序的并行分治Join算法.理论和实验结果证明,基于单关系排序的分治Join算法高于排序合并Join算法.特别是在并行计算环境下,基于单关系排序的并行分治Join算法的效率远远高于排序合并     

On decentralized dynamic pole placement and feedback stabilization

In this paper we study the feedback control problem using an r-channel decentralized dynamic feedback control scheme. We will develop the theory in the behavioral framework. Using this framework we introduce an algebraic parameterization of the space of all possible feedback compensators having a bounded McMillan degree, and we show that this parameterization has the structure of an algebraic variety. We define the pole-placement map for this problem, and we give exact conditions when this map is onto, and almost onto. Finally we provide new necessary and sufficient conditions which guarantee that the set of stabilizable plants is a generic set     

On the glog-normal distribution and its application to the gene expression problem

In this article, we characterized the glog-normal distribution and present a comprehensive treatment of the properties of this model. Specifically, we present the probability density function as well as a graphical analysis of this density, the cumulative distribution function and the moments for this statistical distribution. Additionally, by using likelihood methods, we estimate the parameters, carry out asymptotic inference and discuss influence diagnostics of this model. Finally, we show the usefulness of the glog-normal distribution for modeling gene expression microarray intensity data by means of a real numerical example.     

A framework for establishing trust in Cloud provenance

Provenance is a vital requirement for the success of Clouds, and it is associated with many challenges that are difficult to deal with. In this paper, we explore this area, we identify the problems in current Cloud provenance, we identify the challenges of having trustworthy secure Cloud provenance, and we identify the requirements which could address the identified challenges. We then propose a foundation framework for establishing trust in Cloud provenance. Finally, we draw our research agenda in this direction.     

Mapping standards for home networking

In this study, we apply a step-by-step approach for the identification of standards for home networking. We develop a classification and we use this classification to categorize sixty-four (sets of) standards. By developing this categorization, we have brought order to the chaos of home networking standards.     

基于数据仓库的篮球教练辅助决策支持系统

现在的篮球教练做决策时候还仅仅依靠原先的经验和专业技能,这样的决策有时候往往会产生不好的效果导致球队输球,教练自己下岗.就这个问题笔者认为可以采用设计一个篮球教练辅助决策支持系统的方法给予一定程度的解决,具体在实现该系统时运用数据仓库和数据挖掘技术的方法,然后还以NBA球队火箭队球员在助攻维度的数据进行实验分析.从得出...     

The Dynamification of Modal Dependence Logic

We examine the transitions between sets of possible worlds described by the compositional semantics of Modal Dependence Logic, and we use them as the basis for a dynamic version of this logic. We give a game theoretic semantics, a (compositional) transition semantics and a power game semantics for this new variant of modal Dependence Logic, and we prove their equivalence; and furthermore, we examine a few of the properties of this formalism and show that Modal Dependence Logic can be recovered from it by reasoning in terms of reachability. Then we show how we can generalize this approach to a very general formalism for reasoning about transformations between pointed Kripke models.     

Planar expropriation problem with non-rigid rectangular facilities

In this paper, we introduce the planar expropriation problem with non-rigid rectangular facilities. The facilities considered in this study are two-dimensional facilities of rectangular shape. Moreover, we allow the facility dimensions to be decision variables and introduce the concept of non-rigid facilities. Based on the geometric properties of such facilities, we developed a new formulation for this continuous covering location model which does not require employing distance measures. This model is intended to determine the location and formation of facilities simultaneously. For solving this new model, we proposed a continuous branch-and-bound framework utilizing linear approximations for the tradeoff curve associated with the facility formation alternatives. Further, we developed new problem generation and bounding strategies suitable for our particular problem structure. Computational experience shows that the branch-and-bound procedure we developed performs better than conventional mixed-integer nonlinear programming solvers BARON and SBB for solving this particular location model.     

A relativistic information theory model for general systems Lorentz transformation of organizability and structural entropy†

In a preceding article which appeared in this journal we have modified the classical information theory of Shannon (mainly we defined the concepts of ‘ effective information ’ and ‘ negative information ’) and by this way we have inferred a thermodynamic model for describing general systems. In the present paper we improve this approach by using a relativistic formulation which explicitly describes the basic relativistic feature of the information notion. After a careful analysis of the ‘ internal entropy ’ and ‘ external entropy ’ concepts, we apply these notions to the universe and we show that this latter exhibits a universal constant. We introduce a Riemannian space to describe the dynamics of a system in function of its external entropy and the Lorentz transformation is given. We define the concepts of ‘ organizability ’, ‘ own external entropy ’ and ‘ universe organizability ’. Composition laws for the organizability are given. Lastly we define the relativistic dynamics of general systems and some examples are outlined.     

非平衡类数据分类概述

本文对非平衡类数据分类问题进行了概述。首先在简单介绍非平衡类数据基本概念的基础上,分析了非平衡类数据引起的问题及其导致分类性能下降的原因;然后介绍了目前主要的解决方法,分析了现有处理方法的优缺点;最后讨论了未来的研究方向。     

一类四阶各向异性扩散方程在图像放大中的应用

在文中我们首先分析了进行图像放大时各向异性偏微分方程优于各向同性偏微分方程,随后我们分析了在本文中不同四阶模型的扩散方向.为了消除低阶偏微分方程在处理图像中出现的块状效应的影响,同时保证方程为各向异性扩散,我们构造了两个各向异性的四阶偏微分方程,并且分别从数据和放大图像效果两方面来说明我们给出的模型优于文中提到的其它四个模型.