检索结果-内蒙古大学图书馆

parallel unsupervised Synthetic Aperture Radar image change detection on a graphics processing unit

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS 2013年第2期27卷 109-122页

作者： Zhu, Huming Cao, Yu Zhou, Zhiqiang Gong, Maoguo Jiao, Licheng Xidian Univ Minist Educ China Key Lab Intelligent Percept & Image Understanding Xian 710071 Peoples R China

Change detection is now routinely applied in many application domains, such as damage assessment, environmental monitoring and agricultural surveys. As the number of remote sensing images and the complexity of algorithms rise, the demand for processing power is increasing. In this paper, we propose PLog-FLICM , a parallel algorithm for change detection, which includes two steps: (1) generate the difference image based on the log-ratio operator;(2) detect changes in the difference image by using a modified fuzzy c-means clustering algorithm. PLog-FLICM is implemented on AMD Accelerated parallel Processing SDK based on Open Computing Language. The parallel characteristics and implementation details of the proposed PLog-FLICM algorithm are presented. Experiments on several Synthetic Aperture Radar images demonstrate that the proposed algorithm outperforms other algorithms, and the designed parallel algorithm can greatly reduce the computational time of the change detection algorithm. Furthermore, we investigate the performance portability of PLog-FLICM in the different central processing unit and graphics processing unit platforms. Experimental results show that they have also achieved good parallel performance.

关键词： parallel algorithm graphics processing unit image processing change detection multi-temporal analysis

来源：评论

学校读者我要写书评

暂无评论

parallel globally optimal structure learning of Bayesian networks

引用

JOURNAL OF parallel AND DISTRIBUTED COMPUTING 2013年第8期73卷 1039-1048页

作者： Nikolova, Olga Zola, Jaroslaw Aluru, Srinivas Sage Bionetworks Seattle WA 98109 USA Rutgers State Univ Rutgers Discovery Informat Inst Piscataway NJ 08854 USA Iowa State Univ Dept Elect & Comp Engn Ames IA 50011 USA Indian Inst Technol Dept Comp Sci & Engn Bombay 400076 Maharashtra India

Given n random variables and a set of m observations of each of the n variables, the Bayesian network structure learning problem is to learn a directed acyclic graph (DAG) on the n variables such that the implied joint probability distribution best explains the set of observations. Bayesian networks are widely used in many fields including data mining and computational biology. Globally optimal (exact) structure learning of Bayesian networks takes O(n(2).2(n)) time plus the cost of O(n.2(n)) evaluations of an application-specific scoring function whose run-time is at least linear in m. In this paper, we present a parallel algorithm for exact structure learning of a Bayesian network that is communication-efficient and workoptimal up to O(1/n.2(n)) processors. We further extend this algorithm to the important restricted case of structure learning with bounded node in-degree and investigate the performance gains achievable because of limiting node in-degree. We demonstrate the applicability of our method by implementation on an IBM Blue Gene/P system and an AMD Opteron InfiniBand cluster and present experimental results that characterize run-time behavior with respect to the number of variables, number of observations, and the bound on in-degree. (C) 2013 Elsevier Inc. All rights reserved.

关键词： Bayesian networks Graphical models Machine learning Structure learning parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

Massively parallel Monte Carlo for many-particle simulations on GPUs

引用

JOURNAL OF COMPUTATIONAL PHYSICS 2013年 254卷 27-38页

作者： Anderson, Joshua A. Jankowski, Eric Grubb, Thomas L. Engel, Michael Glotzer, Sharon C. Univ Michigan Dept Chem Engn Ann Arbor MI 48109 USA Univ Michigan Dept Mat Sci & Engn Ann Arbor MI 48109 USA

Current trends in parallel processors call for the design of efficient massively parallel algorithms for scientific computing. parallel algorithms for Monte Carlo simulations of thermodynamic ensembles of particles have received little attention because of the inherent serial nature of the statistical sampling. In this paper, we present a massively parallel method that obeys detailed balance and implement it for a system of hard disks on the GPU. We reproduce results of serial high-precision Monte Carlo runs to verify the method. This is a good test case because the hard disk equation of state over the range where the liquid transforms into the solid is particularly sensitive to small deviations away from the balance conditions. On a Tesla K20, our GPU implementation executes over one billion trial moves per second, which is 148 times faster than on a single Intel Xeon E5540 CPU core, enables 27 times better performance per dollar, and cuts energy usage by a factor of 13. With this improved performance we are able to calculate the equation of state for systems of up to one million hard disks. These large system sizes are required in order to probe the nature of the melting transition, which has been debated for the last forty years. In this paper we present the details of our computational method, and discuss the thermodynamics of hard disks separately in a companion paper. (C) 2013 Elsevier Inc. All rights reserved.

关键词： Monte Carlo parallel algorithm Detailed balance GPGPU CUDA Hard disk system

来源：评论

学校读者我要写书评

暂无评论

parallel computation of determinants of matrices with multivariate polynomial entries

引用

Science China(Information Sciences) 2013年第11期56卷 158-173页

作者： CHEN LiangYu ZENG ZhenBing Shanghai Key Laboratory of Trustworthy Computing East China Normal University

In this paper we present an extension to the work of Bjrck et *** computing the determinants of matrices with univariate or bivariate polynomials as entries to multivariate *** algorithm supports parallel computation and has been implemented on a multi-core cluster computer *** show how to use our approach to calculate two unsolved problems,which arise from computational geometry optimization and electric power engineering,and analyze the time complexity as well as bits complexity.

关键词： determinant interpolation parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

parallel defect-correction algorithms based on finite element discretization for the Navier-Stokes equations

引用

COMPUTERS & FLUIDS 2013年 79卷 200-212页

作者： Shang, Yueqiang Guizhou Normal Univ Sch Math & Comp Sci Guiyang 550001 Peoples R China

Based on a fully overlapping domain decomposition technique and finite element discretization, two parallel defect-correction algorithms for the stationary Navier-Stokes equations with high Reynolds numbers are proposed and investigated. In these algorithms, each processor first solves an artificial viscosity stabilized Navier-Stokes equations by Newton or Picard iterative method, and then diffuses the system in the correction steps where only a linear problem needs to be solved at each step. All the computations are performed in parallel on global composite meshes that are fine around a particular subdomain and coarse elsewhere. The algorithms have low communication complexity. They can yield an approximate solution with an accuracy comparable to that of the standard finite element solution. Numerical tests demonstrated the effectiveness of the algorithms. (C) 2013 Elsevier Ltd. All rights reserved.

关键词： Navier-Stokes equations Finite element Defect-correction method parallel computing parallel algorithm Domain decomposition

来源：评论

学校读者我要写书评

暂无评论

A parallel two-level finite element variational multiscale method for the Navier-Stokes equations

引用

NONLINEAR ANALYSIS-THEORY METHODS & APPLICATIONS 2013年 84卷 103-116页

作者： Shang, Yueqiang Guizhou Normal Univ Sch Math & Comp Sci Guiyang 550001 Peoples R China

A combination method of two-grid discretization approach with a recent finite element variational multiscale algorithm for simulation of the incompressible Navier-Stokes equations is proposed and analyzed. The method consists of a global small-scale nonlinear Navier-Stokes problem on a coarse grid and local linearized residual problems in overlapped fine grid subdomains, where the numerical form of the Navier-Stokes equations on the coarse grid is stabilized by a stabilization term based on two local Gauss integrations at element level and defined by the difference between a consistent and an under-integrated matrix involving the gradient of velocity. By the technical tool of local a priori estimate for the finite element solution, error bounds of the discrete solution are estimated. algorithmic parameter scalings are derived. Numerical tests are also given to verify the theoretical predictions and demonstrate the effectiveness of the method. (C) 2013 Elsevier Ltd. All rights reserved.

关键词： Navier-Stokes equations Finite element Variational multiscale method parallel algorithm Two-grid method

来源：评论

学校读者我要写书评

暂无评论

An effective and efficient parallel approach for random graph generation over GPUs

引用

JOURNAL OF parallel AND DISTRIBUTED COMPUTING 2013年第3期73卷 303-316页

作者： Bressan, Stephane Cuzzocrea, Alfredo Karras, Panagiotis Lu, Xuesong Nobari, Sadegh Heyrani Natl Univ Singapore Singapore 117417 Singapore ICAR CNR I-87036 Cosenza Italy Univ Calabria I-87036 Cosenza Italy

The widespread usage of random graphs has been highlighted in the context of database applications for several years. This because such data structures turn out to be very useful in a large family of database applications ranging from simulation to sampling, from analysis of complex networks to study of randomized algorithms, and so forth. Amongst others, Erdos-Renyi Gamma(v,p) is the most popular model to obtain and manipulate random graphs. Unfortunately, it has been demonstrated that classical algorithms for generating Erdos-Renyi based random graphs do not scale well in large instances and, in addition to this, fail to make use of the parallel processing capabilities of modern hardware. Inspired by this main motivation, in this paper we propose and experimentally assess a novel parallel algorithm for generating random graphs under the Erdos-Renyi model that is designed and implemented in a Graphics Processing Unit (GPU), called PPreZER. We demonstrate the nice amenities due to our solution via a succession of several intermediary algorithms, both sequential and parallel, which show the limitations of classical approaches and the benefits due to the PPreZER algorithm. Finally, our comprehensive experimental assessment and analysis brings to light a relevant average speedup gain of PPreZER over baseline algorithms. (C) 2012 Elsevier Inc. All rights reserved.

关键词： Random graph Random graph generation parallel algorithm GPU

来源：评论

学校读者我要写书评

暂无评论

A very efficient O(n), implicit and parallel method to solve the stream power equation governing fluvial incision and landscape evolution

引用

GEOMORPHOLOGY 2013年 180卷 170-179页

作者： Braun, Jean Willett, Sean D. Univ Grenoble 1 Inst Sci Terre F-38041 Grenoble 9 France CNRS F-38041 Grenoble 9 France ETH Dept Earth Sci Zurich Switzerland

We present a new algorithm to solve the basic stream power equation, which governs channel incision and landscape evolution in many geomorphic settings. The algorithm is highly efficient because computation time increases linearly with the number of points used to discretize the landscape and is ideally suited to parallelization. It is also unconditionally stable because it uses an implicit scheme for the time integration of the landscape evolution equation, which means that large time steps can be used without sacrificing accuracy. In this paper we describe the algorithm and present results that demonstrate its efficiency and accuracy. (C) 2012 Elsevier B.V. All rights reserved.

关键词： Fluvial geomorphology Numerical modeling Stream power equation Routing algorithm Implicit scheme parallel algorithm

来源：评论

学校读者我要写书评

暂无评论

parallel-Data Reconstruction for Limited Views Tomography Sensors by Sinusoidal Hough Transform

引用

IEEE SENSORS JOURNAL 2013年第2期13卷 581-588页

作者： Cantoral-Ceballos, Jose A. Ozanyan, Krikor B. Univ Manchester Sch Elect & Elect Engn Manchester M13 9PL Lancs England

Physical limitations found in industrial environments often restrain imaging for process tomography. When information is collected from sparse sensors, the acquired data is limited in terms of radial and angular sampling of the imaged slice. To overcome this problem, we demonstrate an efficient solution based on the parallel implementation of the sinogram recovery algorithm (SRA) for limited views in its variant based on the calculation of the coordinates of the center of mass (CoMs) of the subject under test, rather than performing the complete sinogram restoration. By introducing a modification in the existing SRA, we achieve high parallelization of each stage, making it ideal for implementation in hardware accelerated systems, such as field programmable gate arrays. The potential to parallelize the SRA is first studied in MATLAB, by processing all data projections concurrently and verifying performance by matching the results from the parallel and sequential implementations. Furthermore, the algorithm is coded in very high speed integrated circuits hardware description language, which is implemented and tested on a Xilinx Virtex 6 board. We report speedups of between three and four orders of magnitude, whereas the errors in CoMs' coordinates are reduced.

关键词： Field programmable gate array (FPGA) Hough transform parallel algorithm sinogram recovery tomography very high speed integrated circuits hardware description language (VHDL)

来源：评论

学校读者我要写书评

暂无评论

parallel numerical solution of incompressible full Navier-Stokes in 3-D Applications to free and mixed convection problems

引用

INTERNATIONAL JOURNAL OF NUMERICAL METHODS FOR HEAT & FLUID FLOW 2013年第4期23卷 553-574页

作者： Unal, V. Ustoglu Yeditepe Univ Dept Phys Istanbul Turkey

Purpose - The purpose of this paper is to develop and test an implicit scheme, accurate to the second order, for solving full Navier-Stokes equations for three dimensional problems, using parallel algorithm. Design/methodology/approach - parallel solution to the 3-D incompressible full Navier-Stokes equations is presented, based on two fractional steps in time and finite element in space. The accuracy of the scheme is second order in both time and space domains. Large time-step sizes, with Courant-Friedrichs-Lewy (CFL) numbers much larger than unity, are taken since the momentum equation is solved implicitly. A fourth order artificial viscosity term is added. In order to stabilize the numerical solution, fourth order artificial viscosity term is used for high Reynolds number flows. The domain decomposition technique is implemented for parallel solution to the problem with matching and non-overlapping sub-domains. It is aimed to study both a 3D free and mixed convection problems using the developed scheme. The segregate solution for temperature field is calibrated by a 3-D free convection problem. Then the flow case where the forced convection is one order of magnitude higher than the free convection is studied. Findings - It is observed that the long time solution to the flow field shows oscillatory behaviour as the Reynolds number of the flow doubled while keeping the ratio of the forced to free convection fixed. The solution using a parallel algorithm gives satisfactory results, in terms of computation time and accuracy, for the natural convection problem in cubic cavity, and, the forced cooling of a room with chilled ceiling having a parabolic geometry as presented at the end. It is observed that doubling the Reynolds number, while keeping all the parameters unchanged, varies the flow behaviour completely. Originality/value - A code previously developed and published by the author only solved momentum equation and studied the velocity field. In this study, full N

关键词： 3-D full Navier-Stokes parallel algorithm Mixed convection problem Implicit parallel solution Domain decomposition Forced-free convection 3D free-convection problem Flow Fluids Convection

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：