一种特殊的H--网孔的性能研究

给出了一种特殊的互联网络模型—H-网孔。它由普通四度网孔的每个结点按规则删去一条边而产生．在最为常见的方形网孔中，其边数减少了25%，平均距离增加了约20％，而直径和普通四度网孔几乎一样．文中还给出了针对H-网孔的广播算法。     

h维墙式环托

定义了h维墙不托hDWT：给定偶数ni≥4,i=1,2,…，h,h维墙式环托hDWT(n1,n2,…，nh)是一个（h 1)-正则的二部图，可由h维普通环托hDOT(n1,n2,…，nh)删除每个节点的h-1条连边而得到。它是节点对称的且有最优容错度h 1,若n1=maxni,则其直么hDOT的直径相同。hDOT可嵌入到hDWT中，其伸张度为3，拥塞度为min{h,4}，因而hDOT上现有的算法都能移植到hDWT上，且有常数的减速比；提出一种简单的自主路由算法，该算法实现hDWT上的任意两个节点间的最短路径。通过把删除的度重新添加到每个节点上，得到增强型hDWT，它比hDOT有小得多的直径，且包含一些有用的互联拓扑结构。h维墙式网孔hDWT可由hDWT翻除所有的回卷边而得到，它的许多性质与hDWT的性质相类似。     

基于Cayley图的六度环绕网络研究

六度网络是一类平面图网络结构,将平面以等边三角形的形式进行分割,包括六度网孔网络和六度环绕网络.六度网孔网络不是规则网络,其边缘节点与内部节点的度不相等.通过对六度网孔网络的边缘节点建立环绕边就形成了规则的六度环绕网络,每个节点的度为6.但是由于环绕边的存在,使得六度环绕网络的通信算法实现复杂,网络直径也非常难于计算.六度环绕网络被证实是一种Cayley图模型,具有良好的对称性.但是基于Cayley图的六度环绕网络的最优路由算法、广播算法还没有得到,该网络模型的具体直径值也是未解问题.针对基于Cayley图的六度环绕网络模型,文中给出了一种简单的最优路由算法和一种基于陪集图理论的广播算法,并给出该网络模型的网络直径确切值.     

地板式三度网孔的性能分析

文章提出了一种新的互连网络模型,它可看作由普通网孔的每个结点删去一条边而产生。文中讨论了这种新式网孔的拓扑性能,并给出了有效的广播算法。     

SAN中的分布式锁机制

总结了当今关于在SAN中的互斥机制的研究,认为将互斥信息完全分布到共享设备和各个互斥节点上是其未来的发展方向．现存的一个分布式锁机制使用双链表组织SAN中的各节点,因此具有节点间传递消息效率低的缺点．为此提出了一种网状结构,并给出了它的相关算法和容错设计以及性能分析．通过模拟实验证实了使用网状结构来组织节点可以大大提高消息传递的效率．网状结构也可以应用于其他需要广播消息但又信息分布的应用中．     

两种GPU上改进的最短路径算法

针对图论中的最短路径问题,提出了两种在GPU上改进的最短路径搜索算法,即针对单源最短路径问题的基于迭代方式且采用原子锁优化的Advanced_Atomics_SSSP算法以及针对所有顶点间最短路径问题的采用二叉堆优化的Heap_APSP算法。将两种算法应用到美国公路网图和节点的度均为6的普通图中,通过对算法的测试表明,Advanced_Atomics_SSSP算法的性能依赖于节点的度数,当节点的度数大于6时加速效果明显,当节点度数为1～3时加速效果不明显;而Heap_APSP可以达到46～56倍的加速比,且加速性能不受节点度的影响。     

基于相似度的三元社团合并算法*

针对使用相似度测量进行社团划分时可能出现的判断冲突问题,提出了一种基于相似度的三元社团合并算法。首先通过对相似度阈值的选取,筛选网络中不同的三元社团,并将其作为社团合并的基本元素,通过社团相似度将其合并。然后将剩余节点和孤立三元社团分别按照节点从属度和三元社团从属度划分到相应社团。最后通过在人工合成网络和真实世界网络上进行实验测试,结果表明用本文算法可以准确高效的将网络中的节点划分到相应的社团。     

利用视频摘要的方式产生多条视频流(英文)

在无线Ad Hoc网络上的视频传输要比在其他网络的视频传输更富于挑战性。由于无线Ad Hoc网络中的节点可以用任意方式动态地连接到一起,通过给定的网孔状的拓扑,在网络传输的源端节点和终端节点之间建立起不止一条的多路径是完全可能的。为在多路径传输中结合多条流的编码方式,开发了一种基于率歪曲优化的算法,进而把一条未加工的视频流分成多条流。实验证明,该算法在最大化多路径传输的效率方面是非常有效的。     

一种协调的科技文献分类方法

科技文献之间的相互引证关系反映了一种科学交流活动,显示了科学文献之间(甚至是学科之间)的内在联系,而通过追溯文献之间的这种关系,可以改善和提高传统的基于内容的科技文献的分类的方法。论文利用有相互引证关系,有同引关系,以及有耦合关系的两篇文献一般是属于同一类的这一特点,提出了文献之间的引用相似度,同引相似度,耦合相似度这三个概念,再利用这三个概念生成了文献之间的“结构相似度”,并将它用于K-NN分类法中得出一种基于结构的分类法。最后,论文将这种基于结构的分类法和基于内容的NaveBayes分类法结合起来提出了一种新的协调分类法。     

BB图的枚举特征

在并发程序复杂性度量研究中，我们曾定义了一种B图，用以作为Ada并发程序中一种会合关系的模型。我们将B图推广到BB图。n节点BB图是在n个节点的串联树上再添加若干条边，其约束条件是：每个节点的入度不大于二，每个节点的出度也不大于二。给出BB图的若干若干枚举特征，指出同第二类Stiding数的密切关系。     

Honeycomb networks: Topological properties and communicationalgorithms

The honeycomb mesh, based on hexagonal plane tessellation, is considered as a multiprocessor interconnection network. A honeycomb mesh network with n nodes has degree 3 and diameter ≈1.63√n-1, which is 25 percent smaller degree and 18.5 percent smaller diameter than the mesh-connected computer with approximately the same number of nodes. Vertex and edge symmetric honeycomb torus network is obtained by adding wraparound edges to the honeycomb mesh. The network cost, defined as the product of degree and diameter, is better for honeycomb networks than for the two other families based on square (mesh-connected computers and tori) and triangular (hexagonal meshes and tori) tessellations. A convenient addressing scheme for nodes is introduced which provides simple computation of shortest paths and the diameter. Simple and optimal (in the number of required communication steps) routing, broadcasting, and semigroup computation algorithms are developed. The average distance in honeycomb torus with n nodes is proved to be approximately 0.54√n. In addition to honeycomb meshes bounded by a regular hexagon, we consider also honeycomb networks with rhombus and rectangle as the bounding polygons     

Using sequential NETGEN as a component for a parallel mesh generator

A parallel tetrahedral mesh generator is developed using the existing sequential NETGEN mesh generator. Mesh generation algorithms developed decompose the geometry into multiple sub-geometries sequentially on a master node and then mesh each sub-geometry in parallel on multiple processors. Two methods are implemented. The first decomposes the geometry and produces conforming surface sub-meshes from which volume meshes can be generated in parallel. A second refinement based method also makes use of the CAD geometry information. A scalable mesh migration algorithm that utilizes “owner updates” rule is implemented. Results show that using the refinement based method, a mesh with a billion elements can be generated in about a minute.     

三维实体仿真建模的网格自动生成方法

有限元网格模型的生成与几何拓扑特征和力学特性有直接关系。建立网格模型时，为了更真实地反映原几何形体的特征，在小特征尺寸或曲率较大等局部区域网格应加密剖分；为提高有限元分析精度和效率，在待分析的开口、裂纹、几何突变、外载、约束等具有应力集中力学特性的局部区域，网格应加密剖分。为此，该文提出了基于几何特征和物理特性相结合的网格自动生成方法。该方法既能有效地描述几何形体，又能实现应力集中区域的网格局部加密及粗细网格的均匀过渡。实例表明本方法实用性强、效果良好。     

A CPU–GPU framework for optimizing the quality of large meshes

The automatic generation of 3D finite element meshes (FEM) is still a bottleneck for the simulation of large fluid dynamic problems. Although today there are several algorithms that can generate good meshes without user intervention, in cases where the geometry changes during the calculation and thousands of meshes must be constructed, the computational cost of this process can exceed the cost of the FEM. There has been a lot of work in FEM parallelization and the algorithms work well in different parallel architectures, but at present there has not been much success in the parallelization of mesh generation methods. This paper will present a massive parallelization scheme for re-meshing with tetrahedral elements using the local modification algorithm. This method is frequently used to improve the quality of elements once the mesh has been generated, but we will show it can also be applied as a regeneration process, starting with the distorted and invalid mesh of the previous step. The parallelization is carried out using OpenCL and OpenMP in order to test the method in a multiple CPU architecture and also in Graphics Processing Units (GPUs). Finally we present the speedup and quality results obtained in meshes with hundreds of thousands of elements and different parallel APIs.     

基于逆3 细分的渐进网格生成算法研究

提出一种基于逆3 细分的渐进网格生成算法,用于解决图形的快速传输和显示问 题。算法的基本思路是：将细密网格通过边折叠操作得到简化网格,以细分极限点逼近原始网 格为准则进行网格调整,采用3 细分得到高密度网格,调整后进行逆3 细分,即逐层次删除 部分顶点,生成用于重构渐进网格模型的基网格,并记录每层删除顶点在采用本层表示时相对 于细分计算位置的几何调整量。3 细分过程中三角片数量增长速度较慢,采用逆3 细分利于 生成多层次的渐进网格,经实例验证,逆3 细分生成渐进网格的效果能满足快速、多分辨率显 示要求。     

扫掠法有限元网格生成方法

为了提高有限元网格的生成质量,扫掠法生成六面体网格过程中内部节点定位成为关键一步,在研究复杂扫掠体六面体有限元网格生成算法过程中,提出了一种基于扫掠法的六面体网格生成算法,算法利用源曲面已经划分好的网格和连接曲面的结构化网格,用仿射映射逐层投影,生成目标曲面,提出基于Roca算法的内部节点定位的新算法,运用由外向内推进的波前法思想,生成全部的六面体网格。通过实例表明,该算法快速,稳定,可靠,可处理大量复杂2.5维实体六面体网格生成问题。     

Broadcasting in All-Output-Port Meshes of Trees with Distance-Insensitive Switching

Meshes of trees are hybrids of meshes and trees with outstanding properties, namely small degree and diameter and large bisection width. Moreover, they are known to be area universal, i.e., they can simulate any network with the same wire area with only a polylogarithmic slowdown. Meshes of trees are known to outperform meshes in execution of algorithms with local communication patterns, e.g., sorting, for which distance-sensitive switching, such as store-and-forward, suffices. Nowadays, parallel machines use distance-insensitive, e.g., wormhole, switching. A challenging problem is to design optimal or efficient algorithms for one-to-all broadcast in all-output-port networks with distance-insensitive switching, since the lower bound on the number of rounds, log_Δ+1N, where Δ is the node degree, is very strict. This problem has been solved for tori and hypercubes quite recently. In this paper, we present nearly optimal algorithms for one-to-all broadcast in both square and rectangular 2-D meshes of trees and cube 3-D meshes of trees. The algorithms need at most one round more than the trivial lower bound. We also show requirements for deadlock-free execution of the algorithms. Meshes of trees are not node-symmetric, they are not even regular. This paper shows that, in contrast to meshes, the irregularity is not an obstacle for designing efficient schemes for such an intensive communication pattern, as the all-output-port broadcast.     

Reliable Whisker Weaving via Curve Contraction

Whisker Weaving is an advancing front algorithm for all-hexahedral mesh generation. It uses global information derived from grouping the mesh dual into surfaces, the STC, to construct the connectivity of the mesh, then positions the nodes afterwards. Currently, we are able to reliably generate hexahedral meshes for complicated geometries and surface meshes. However, the surface mesh must be modified locally. Also, in large, highly-unstructured meshes, there are usually isolated regions where hex quality is poor. Reliability has been achieved by using new, provable curvecontraction algorithms to sequence the advancing front process. We have also demonstrated that sheet moving can remove certain types of invalid connectivity.