19th International Symposium on High Performance Computing Systems and Applications (HPCS'05)

15-18 May 2005

Filter Results

Displaying Results 1 - 25 of 59
  • Proceedings. 19th International Symposium on High Performance Computing Systems and Applications

    Publication Year: 2005
    Request permission for commercial reuse | PDF file iconPDF (280 KB)
    Freely Available from IEEE
  • 19th International Symposium on High Performance Computing Systems and Applications - Title Page

    Publication Year: 2005, Page(s):i - iii
    Request permission for commercial reuse | PDF file iconPDF (36 KB)
    Freely Available from IEEE
  • 19th International Symposium on High Performance Computing Systems and Applications - Copyright Page

    Publication Year: 2005, Page(s): iv
    Request permission for commercial reuse | PDF file iconPDF (47 KB)
    Freely Available from IEEE
  • 19th International Symposium on High Performance Computing Systems and Applications - Table of contents

    Publication Year: 2005, Page(s):v - viii
    Request permission for commercial reuse | PDF file iconPDF (40 KB)
    Freely Available from IEEE
  • Preface

    Publication Year: 2005, Page(s):ix - x
    Request permission for commercial reuse | PDF file iconPDF (21 KB) | HTML iconHTML
    Freely Available from IEEE
  • Steering Committee

    Publication Year: 2005, Page(s): xi
    Request permission for commercial reuse | PDF file iconPDF (17 KB)
    Freely Available from IEEE
  • Scientific Committee

    Publication Year: 2005, Page(s): xi
    Request permission for commercial reuse | PDF file iconPDF (17 KB)
    Freely Available from IEEE
  • Local Organizing Committee

    Publication Year: 2005, Page(s): xii
    Request permission for commercial reuse | PDF file iconPDF (14 KB)
    Freely Available from IEEE
  • Distributed artificial neural network architectures

    Publication Year: 2005, Page(s):2 - 10
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (608 KB) | HTML iconHTML

    The computational cost of training artificial neural network (ANN) algorithms limits the use of large systems capable of processing complex problems. Implementing ANNs on a parallel or distributed platform to improve performance is therefore desirable. This work illustrates a method to predict and evaluate the performance of distributed ANN algorithms by analyzing the performance of the comparativ... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Benchmarking of a 3D, unstructured, finite volume code for incompressible Navier-Stokes equation on a cluster of distributed-memory computers

    Publication Year: 2005, Page(s):11 - 16
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (3552 KB) | HTML iconHTML

    The parallelization and performance of an implicit, unstructured, time-dependent computational fluid dynamics (CFD) code is described. Parallelization of the code is done within the PETSc framework using a single-program-multiple-data (SPMD) message passing model. The parallel code is shown to scale linearly within the limit of the available number of processors. A dynamic convergence criteria (DC... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A meta-software system for the discovery of Hadamard matrices

    Publication Year: 2005, Page(s):17 - 23
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (120 KB) | HTML iconHTML

    The purpose of this paper is to describe some of the technical and engineering concerns which define the search for Hadamard matrices through the creation of a software system which is correct, efficient and extensible. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Quintic spline based computational scheme for singularly perturbed convection-diffusion problems

    Publication Year: 2005, Page(s):24 - 28
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (144 KB) | HTML iconHTML

    In this article, we propose a fourth-order difference scheme using quintic spline for singularly perturbed convection-diffusion Dirichlet boundary-value problems. Theoretical error estimates are derived for the method. Computational efficiency and accuracy are verified through numerical examples. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The study and implementation of the multigrid algorithm for the 3-D transport difference equation

    Publication Year: 2005, Page(s):29 - 31
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (623 KB) | HTML iconHTML

    In order to accelerate the iterative convergence, the multigrid algorithm for the 3D time-dependent neutron transport difference equation is studied in this paper. The detail description for this algorithm and the processes of the implementation are given, and the comparison and conclusion for the numerical results are listed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The analysis of large order bessel functions in gravitational wave signals from pulsars

    Publication Year: 2005, Page(s):34 - 41
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (392 KB) | HTML iconHTML

    In this work, we present the analytic treatment of the large order Bessel functions that arise in the Fourier transform. (FT) of the gravitational wave (GW) signal from a pulsar. We outline several strategies, which employs asymptotic expansions in evaluation of such Bessel functions, which also happen to have large argument. Large order Bessel functions also arise in the Peters-Mathews model of b... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The effects of stellar winds on the envelope masses of cooling low-mass dwarfs

    Publication Year: 2005, Page(s):42 - 48
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (400 KB) | HTML iconHTML

    The determination of the true ages of binary millisecond pulsars is critical to our understanding of their spin-down evolution. The only independent way to estimate these ages is to observe the temperatures of their binary companions and to infer their ages based on the results of detailed stellar evolution computations. In this paper we present recent calculations that quantify the effects of (pa... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High performance derivative-free optimization applied to biomedical image registration

    Publication Year: 2005, Page(s):50 - 56
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (480 KB) | HTML iconHTML

    Optimization of a similarity metric is an essential component in most medical image registration approaches based on image intensities. In this paper, two new, deterministic, derivative-free optimization algorithms are parallelized and adapted for image registration. DIRECT (dividing rectangles) is a global technique for linearly bounded problems, and the multidirectional search (MOS) is a local m... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A customizable component for low-level communication software

    Publication Year: 2005, Page(s):58 - 64
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (168 KB) | HTML iconHTML

    This paper discusses the design of a customizable component that encapsulates a raw baseline communication architecture and a set of lightweight protocols. These protocols can act in specific points of the baseline architecture's algorithm in order tailor communication or provide new services according to the requirements of specific classes of applications. An implementation of the described comp... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Hpcbench - a Linux-based network benchmark for high performance networks

    Publication Year: 2005, Page(s):65 - 71
    Cited by:  Papers (4)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (192 KB) | HTML iconHTML

    In recent years, Linux-based clusters have become more prevalent as a basis for high performance computing (HPC) systems. Network performance analysis is crucial to the management and administration of such clusters. To assist in this process, we developed Hpcbench to measure UDP, TCP and MPl communications over high performance networks. Hpcbench records and tracks experiment results and system s... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A replacement policy to save energy for data cache

    Publication Year: 2005, Page(s):72 - 75
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (90 KB) | HTML iconHTML

    For a current processor, it has been reported that more than 50% of power has been consumed by an on-chip cache memory. To reduce power consumption for the cache memory, it is required to develop an effective replacement policy and mapping function, which would decide the cache memory accesses. In this paper, we introduce an ELRU-SEQ replacement policy with skewed mapping function to save energies... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Applying fault-tolerant solutions of circulant graphs to meshes and hypercubes

    Publication Year: 2005, Page(s):76 - 82
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (160 KB) | HTML iconHTML

    Many important architectures such as rings, meshes and hypercubes can be modeled as circulant graphs. As a result, circulant graphs have received a lot of attention, and a new method was developed for designing fault-tolerant solutions for them. We review this method in this paper, and examine its applications to the design of fault-tolerant solutions for meshes and hypercubes. Our results indicat... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • B-rep based parallel machining simulation

    Publication Year: 2005, Page(s):83 - 89
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (272 KB) | HTML iconHTML

    Boundary-representation (B-rep) solid modeller based machining simulations are employed in industry to avoid potential manufacturing failures and improve machining efficiency before machining begins. The B-rep solid modeller maintains the state of the in-process stock exactly. However, the application of B-rep solid modellers to machining simulations is hindered by the complexity growth that occur... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Exploiting multithreaded programming on cluster architectures

    Publication Year: 2005, Page(s):90 - 96
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (168 KB) | HTML iconHTML

    The development of programs for parallel architectures focusing to achieve high indices of performance is a hard task. The programmer must deal with two classes of problems, one related to describing the application concurrency and data dependencies, and other related to exploiting efficiently the processing resources of the architecture, in order to obtain an efficient implementation for a specif... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Exploring parallel programming knowledge in the novice

    Publication Year: 2005, Page(s):97 - 102
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (67 KB) | HTML iconHTML

    Parallel programming is a difficult task for programmers. Simplifying the learning of this job first requires that we understand what needs to be taught. Evaluating the novice and expert parallel programmer for their knowledge framework can aid in the transition. This exploratory study examines novices' assumptions about the difficulty of parallel programming. Novices in this pilot study were capa... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A fully parallel and scalable implementation of a Hopfield neural network on the SHARC-net supercomputer

    Publication Year: 2005, Page(s):103 - 109
    Cited by:  Papers (1)
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (160 KB) | HTML iconHTML

    Artificial neural networks (ANN) are an established area of artificial intelligence (AI) and computer science. ANNs have been used in a number of ways for research and industrial projects. However, despite ANN research spanning many years, the typical implementation is a single threaded programming model. This paper presents a fully parallel implementation of a Hopfield neural network using a supe... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The HPCVL working template: a tool for high-performance programming

    Publication Year: 2005, Page(s):110 - 116
    Request permission for commercial reuse | Click to expandAbstract | PDF file iconPDF (112 KB) | HTML iconHTML

    The HPCVL working template is a set of scripts and libraries that facilitates programming for high-performance machines. This is achieved by providing three main functionalities: (1) the maintenance of multiple versions with only a single original source, (2) an automatic relative debugger that is suitable to use one version of the code (often the serial one) to check the correctness of another (f... View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.