1,260 Matching Results

Search Results

Advanced search parameters have been applied.

in subjects: "Parallel Processing"

open access

Evaluation of a Cray performance tool using a large hydrodynamics code

This paper will discuss one of these automatic tools that has been developed recently by Cray Research, Inc. for use on its parallel supercomputer. The tool is called ATEXPERT; when used in conjunction with the Cray Fortran compiling system, CF77, it produces a parallelized version of a code based on loop-level parallelism, plus information to enable the programmer to optimize the parallelized code and improve performance. The information obtained through the use of the tool is presented in an … more

Date: January 1, 1992

Creator: Lord, K.M. (Cray Research, Inc., Eagan, MN (United States)) & Simmons, M.L. (Los Alamos National Lab., NM (United States))

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

ADIFOR: Automatic differentiation in a source translator environment

The numerical methods employed in the solution of many scientific computing problems require the computation of derivatives of a function f: R{sup n} {yields} R{sup m}. ADIFOR (Automatic Differentiation in FORtran) is a source transformation tool that accepts Fortran 77 code for the computation of a function and writes portable Fortran 77 code for the computation of the derivatives. In contrast to previous approaches, ADIFOR views automatic differentiation as a source transformation problem and… more

Date: January 1, 1992

Creator: Bischof, C.; Corliss, G.; Griewank, A. (Argonne National Lab., IL (United States)) & Carle, A. (Rice Univ., Houston, TX (United States). Center for Research on Parallel Computation)

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

ADIFOR: Fortran source translation for efficient derivatives

The numerical methods employed in the solution of many scientific computing problems require the computation of derivatives of a function f: R{sup n} {yields} R{sup m}. Both the accuracy and the computational requirements of the derivative computation are usually of critical importance for the robustness and speed of the numerical method. ADIFOR (Automatic Differentiation In FORtran) is a source translation tool implemented using the data abstractions and program analysis capabilities of the Pa… more

Date: January 1, 1992

Creator: Bischof, C.; Corliss, G.; Griewank, A.; Hovland, P. (Argonne National Lab., IL (United States)) & Carle, A. (Rice Univ., Houston, TX (United States). Center for Research on Parallel Computation)

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

An assessment of the ModSim/TWOS parallel simulation environment

The Time Warp Operating System (TWOS) has been the focus of significant research in parallel, discrete-event simulation (PDES). A new language, ModSim, has been developed for use in conjunction with TWOS. The coupling of ModSim and TWOS is an attempt to address the development of large-scale, complex, discrete-event simulation models for parallel execution. The approach, simply stated, is to provide a high-level simulation-language that embodies well-known software engineering principles combin… more

Date: January 1, 1991

Creator: Rich, D.O. & Michelsen, R.E.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

On finding minimum-diameter clique trees

It is well-known that any chordal graph can be represented as a clique tree (acyclic hypergraph, join tree). Since some chordal graphs have many distinct clique tree representations, it is interesting to consider which one is most desirable under various circumstances. A clique tree of minimum diameter (or height) is sometimes a natural candidate when choosing clique trees to be processed in a parallel computing environment. This paper introduces a linear time algorithm for computing a minimum-… more

Date: August 1, 1991

Creator: Blair, J.R.S. (Tennessee Univ., Knoxville, TN (United States). Dept. of Computer Science) & Peyton, B.W. (Oak Ridge National Lab., TN (United States))

Item Type: Report

Partner: UNT Libraries Government Documents Department

open access

A domain decomposition algorithm for solving large elliptic problems

AN algorithm which efficiently solves large systems of equations arising from the discretization of a single second-order elliptic partial differential equation is discussed. The global domain is partitioned into not necessarily disjoint subdomains which are traversed using the Schwarz Alternating Procedure. On each subdomain the multigrid method is used to advance the solution. The algorithm has the potential to decrease solution time when data is stored across multiple levels of a memory hier… more

Date: January 1, 1991

Creator: Nolan, M.P.

Item Type: Report

Partner: UNT Libraries Government Documents Department

open access

Front-end data processing the SLD data acquisition system

The data acquisition system for the SLD detector will make extensive use of parallel at the front-end level. Fastbus acquisition modules are being built with powerful processing capabilities for calibration, data reduction and further pre-processing of the large amount of analog data handled by each module. This paper describes the read-out electronics chain and data pre-processing system adapted for most of the detector channels, exemplified by the central drift chamber waveform digitization a… more

Date: July 1, 1986

Creator: Nielsen, B. S.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Parallel supercomputing: Advanced methods, algorithms and software for large-scale problems

Research has continued with excellent progress and new results on methodology and algorithms. We have also made supporting benchmark application studies on representative parallel computing architectures. Results from these research studies have been reported at scientific meetings, as technical reports and as journal publications. A list of pertinent presentations and publications is attached. The work on parallel element-by-element techniques and domain decomposition schemes has developed wel… more

Date: April 1, 1992

Creator: Carey, G.F. & Young, D.M.

Item Type: Report

Partner: UNT Libraries Government Documents Department

open access

Progress Report: Network Based High Performance Concurrent Computing

The overall objectives of this project are to investigate research issues pertaining to programming tools and efficiency issues in network based concurrent computing systems. The basis for these efforts is the PVM project that evolved during my visits to Oak Ridge Laboratories under the DOE Faculty Research Participation program; I continue to collaborate with researchers at Oak Ridge on some portions of the project.

Date: 1991~

Creator: Sunderam, V. S.

Item Type: Report

Partner: UNT Libraries Government Documents Department

open access

Debugging in a parallel environment

This paper describes the preliminary results of a project investigating approaches to dynamic debugging in parallel processing systems. Debugging programs in a multiprocessing environment is particularly difficult because of potential errors in synchronization of tasks, data dependencies, sharing of data among tasks, and irreproducibility of specific machine instruction sequences from one job to the next. The basic methodology involved in predicate-based debuggers is given as well as other desi… more

Date: January 1, 1985

Creator: Wasserman, H.J. & Griffin, J.H.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Toward a science of parallel computation

The evolution of parallel processing over the past several decades can be viewed as the development of a new scientific discipline. Parallel processing has been, and is, undergoing the same evolutionary stages that are common to the development of scientific disciplines in general: exploration, focusing, and maturity. That parallel processing is not yet a science can readily be appreciated by its lack of some of the characteristics typical of mature sciences, such as prescriptive terminology, c… more

Date: January 1, 1986

Creator: Worlton, W.J.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Some computational challenges of developing efficient parallel algorithms for data-dependent computations in thermal-hydraulics supercomputer applications

The Transient Reactor Analysis Code (TRAC), which features a two- fluid treatment of thermal-hydraulics, is designed to model transients in water reactors and related facilities. One of the major computational costs associated with TRAC and similar codes is calculating constitutive coefficients. Although the formulations for these coefficients are local the costs are flow-regime- or data-dependent; i.e., the computations needed for a given spatial node often vary widely as a function of time. C… more

Date: January 1, 1992

Creator: Woodruff, S.B.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Writing Parallel, Discrete-Event Simulations in Modsim: Insight and Experience

The Time Warp Operating System (TWOS) has been the focus of much research in parallel simulation. A new language, called ModSim, has been developed for use in conjunction with TWOS. The coupling of ModSim and TWOS provides a tool to construct large, complex simulation models that will run on several parallel and distributed computer systems. As part of the Griffin Project'' underway here at Los Alamos National Laboratory, there is strong interest in assessing the coupling of ModSim and TWOS fro… more

Date: September 11, 1989

Creator: Rich, D.O. & Michelsen, R.E.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Mathematical and numerical models to achieve high speed with special-purpose parallel processors

One simulation facility that has been developed is the BNL Plant Analyzer, currently set up for BWR plant simulations at up to seven times faster than real-time process speeds. The principal hardware components of the BNL Plant Analyzer are two units of special-purpose parallel processors, the AD10 of Applied Dynamics International and a PDP-11/34 host computer. The AD10 is specifically designed for time-critical system simulations, utilizing the modern parallel processing technology with pipel… more

Date: July 1, 1986

Creator: Cheng, H.S.; Wulff, W. & Mallen, A.N.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Information technology resources assessment

This year's Information Technology Resources Assessment (ITRA) is something of a departure from traditional practice. Past assessments have concentrated on developments in fundamental technology, particularly with respect to hardware. They form an impressive chronicle of decreasing cycle times, increasing densities, decreasing costs (or, equivalently, increasing capacity and capability per dollar spent), and new system architectures, with a leavening of operating systems and languages. Past ass… more

Date: January 1, 1992

Creator: Stevens, D.F. (ed.)

Item Type: Report

Partner: UNT Libraries Government Documents Department

open access

A parallel block cyclic reduction algorithm for the fast solution of elliptic equations

This paper presents an adaptation of the Block Cyclic Reduction (BCR) algorithm for a multi-vector processor. The main bottleneck of BCR lies in the solution of linear systems whose coefficient matrix is the product of tridiagonal matrices. This bottleneck is handled by expressing the rational function corresponding to the inverse of this product as a sum of elementary fractions. As a result the solution of this system leads to parallel solutions of tridiagonal systems. Numerical experiments pe… more

Date: April 1, 1987

Creator: Gallopoulos, E. & Saad, Y.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Progress in the development of a numerical tokamak

This report discusses the numerical tokamak; plasma simulation on magnetohydrodynamic time scales; and plasma simulation in real geometries. (LSP)

Date: January 1, 1991

Creator: Brackbill, J.U.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Matrix factorization on a hypercube multiprocessor

This paper is concerned with parallel algorithms for matrix factorization on distributed-memory, message-passing multiprocessors, with special emphasis on the hypercube. Both Cholesky factorization of symmetric positive definite matrices and LU factorization of nonsymmetric matrices using partial pivoting are considered. The use of the resulting triangular factors to solve systems of linear equations by forward and back substitutions is also considered. Efficiencies of various parallel computat… more

Date: August 1, 1985

Creator: Geist, G. A. & Heath, M. T.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Modeling node bandwidth limits and their effects on vector combining algorithms

Each node in a message-passing multicomputer typically has several communication links. However, the maximum aggregate communication speed of a node is often less than the sum of its individual link speeds. Such computers are called node bandwidth limited (NBL). The NBL constraint is important when choosing algorithms because it can change the relative performance of different algorithms that accomplish the same task. This paper introduces a model of communication performance for NBL computers … more

Date: January 13, 1992

Creator: Littlefield, R.J.

Item Type: Report

Partner: UNT Libraries Government Documents Department

open access

Programming a massively parallel, computation universal system: static behavior

In previous work by the authors, the ''optimum finding'' properties of Hopfield neural nets were applied to the nets themselves to create a ''neural compiler.'' This was done in such a way that the problem of programming the attractors of one neural net (called the Slave net) was expressed as an optimization problem that was in turn solved by a second neural net (the Master net). In this series of papers that approach is extended to programming nets that contain interneurons (sometimes called '… more

Date: January 1, 1986

Creator: Lapedes, A. & Farber, R.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

On the abstracted dataflow complexity of Fast Fourier Transforms

In this paper we develop and analyze the simulated performance of codes for the Fast Fourier Transform written in If and targeted for execution on Motorola's dataflow machine Monsoon. The FFT application is of interest because of its computational parallelism, its requirement for global communications, and its array element data dependences. We use the parallel profiling simulator Id World to study the dataflow performance of various implementations. Our approach is comparative. We study two ap… more

Date: January 1, 1992

Creator: Boehm, A.P.W. (Colorado State Univ., Fort Collins, CO (United States). Dept. of Computer Science); Hiromoto, R.E.; Kelly, K.A. (Los Alamos National Lab., NM (United States)) & Ashley, J.M. (Indiana Univ., Bloomington, IN (United States). Dept. of Computer Science)

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Empirical results of a hybrid Monte Carlo method for the solution of Poisson's equation

The application of Monte Carlo techniques have considerable importance in solving computational problems that exhibit complex physical interactions and irregular spatial geometries. Yet equally important is the large computational parallelism that is inherent in these methods. With the current interest and emphasis on massively parallel computer systems, Monte Carlo methods would seem to be likely candidate for parallel implementation. However, parallelism in itself is only one metric for paral… more

Date: January 1, 1991

Creator: Hiromoto, R.E. & Brickner, R.G.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Thoughts on software and computing

This talk has three distinct parts. The first two parts are on vector and parallel processing and their success, or lack thereof for HEP. The third part is an analysis on the software situation in HEP. These topics have been chosen because of the frequency with which they are discussed in the hallways of our laboratories and institutions. This review looks at the field from a particular point of view: that of an experimental physicist working with a large detector at a collider and, in addition… more

Date: March 1, 1988

Creator: Kunz, P. F.

Item Type: Article

Partner: UNT Libraries Government Documents Department

open access

Information dynamics of self-programmable matter

Using the simple observation that programs are identical to data, programs alter data, and thus programs alter programs, we have constructed a self-programming system based on a parallel von Neumann architecture. This system has the same fundamental property as living systems have: the ability to evolve new properties. We demonstrate how this constructive dynamical system is able to develop complex cooperative structures with adaptive responses to external perturbations. The experiments with th… more

Date: January 1, 1991

Creator: Knudsen, C.; Feldberg, R. (Danmarks Tekniske Hoejskole, Lyngby (Denmark)) & Rasmussen, S. (Los Alamos National Lab., NM (United States))

Item Type: Article

Partner: UNT Libraries Government Documents Department