Interner Bericht

On Solving the McConnell Equations in Biochemistry (2009)

Forschungszentrum Jülich Gmbh, Interner Bericht, Johannes Grotendorst, Paul Jansen, Siegfried M. Schoberth, Johannes Grotendorst, ...

In recent years, in vivo nuclear magnetic resonance (NMR) spectroscopy has allowed to measure rate constants of transport and diffusion across living cell membranes [1-2]. One of the authors (S.M.S.)...

QCD on Parallel Computers at the HLRZ Supercomputing Center (2009)

Forschungszentrum Jülich Gmbh, Interner Bericht, Norbert Attig, Norbert Attig

Large-scale simulations of Quantum Chromodynamics (QCD) which were performed on different parallel high-end supercomputers at the HLRZ supercomputing center, are introduced. Emphasis is laid on the...

Compiling SVM-Fortran for the Intel Paragon XP/S (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rudolf Berrendorf, Michael Gerndt

SVM-Fortran is a language designed to program highly parallel systems with a global address space. A compiler for SVM-Fortran is described which generates code for parallel machines; our current...

Performance Characteristics for OpenMP Constructs on Different Parallel Computer Architectures Abstract (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rudolf Berrendorf, Guido Nieken, Rudolf Berrendorf, Guido Nieken

OpenMP is emerging as a quasi-standard for shared memory parallel programming on small SMP-systems. To serve as a common programming interface in shared memory parallel programming, scalability to a...

Conjugate Gradient and Lanczos Methods for Sparse Matrices on Distributed Memory Multiprocessors (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Achim Basermann, A. Basermann

Conjugate gradient methods to solve sparse systems of linear equations and Lanczos algorithms for sparse symmetric eigenvalue problems play an important role in numerical methods for solving...

Automatic Performance Analysis for Shared Virtual Memory Systems (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Michael Gerndt, Andreas Krumme, M. Gerndt, A. Krumme

Programming distributed memory multiprocessors requires program parallelization as well as program optimization with respect to data locality. SVM-Fortran is a programming language for shared virtual...

on massively parallel machines (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Achim Basermann, Björn Reichel, Christof Schelthoff, A. Basermann, ...

Preconditioned CG methods for sparse matrices on massively parallel machines Dieser Bericht wurde zur Publikation eingereicht.

Interference Graphs of Programs in SSA-form (2008)

Interner Bericht, Sebastian Hack

Register allocation is the task of mapping the variables in a program to processor registers. This problem is often reduced to coloring the so-called interference graph which the compiler computes...

Abstract (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Properties Asl, Thomas Fahringer, Letzte Änderung, Thomas Fahringer, ...

Performance analysis is an important step in tuning performance critical applications. It is a cyclic process of measuring and analyzing performance data which is driven by the programmers hypotheses...

(Betriebssysteme) Fine-grained Parallel Computation in the PM/OM Object Model (2008)

Mathematische Maschinen, Jan-peter Richter, Jürgen Kleinöder, Januar Tr-i, Interner Bericht, Jan-peter Richter, ...

In this paper a novel computation model based on fine-grained objects and asynchronous procedure invocation is presented. Based on a set of only a few basic definitions it can be used to write well...

Modelling of Parallel Processing Tasks by Combinatorial Designs (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Heribert C. Burg, Heribert C. Burg

Combinatorial Designs, also known as experimental designs, historically first were used in agriculture. With the advent of electronic computers together with the significant advancement of...

VAMPIR: Visualization and Analysis of MPI Resources”. http://www.pallas.de/pages/vampir.htm (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Wolfgang E. Nagel, Alfred Arnold, Michael Weber, W. E. Nagel, ...

Performance analysis most often is based on the detailed knowledge of program behavior. One option to get this information is tracing. Based on the research tool PARvis, the visualization environment...

PCL- The Performance Counter Library: A Common Interface to Access Hardware Performance Counters on Microprocessors (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rudolf Berrendorf, Heinz Ziegler, Rudolf Berrendorf, Heinz Ziegler

ii A performance counter is that part of a microprocessor that measures and gathers performance-relevant events on the microprocessor. The number and type of available events differ significantly...

Work Package 2 Common Interfaces and Integration Techniques (2008)

Forschungszentrum Jülich Gmbh, Interner Bericht, Bernd Mohr, Jesper Larsson Träff, Bernd Mohr, Forschungszentrum Jülich, ...

Automatic performance tools must of course be tested as to whether they perform their task correctly. Because performance tools are meta-programs, tool testing is more complex than ordinary program...

Die Gaststudenten und ihre Betreuer waren: (2008)

Interner Bericht, Matthias Bolten (hrsg, Jülich Supercomputing Centre, Interner Bericht, Matthias Bolten (hrsg, ...

Entsprechend dem fächerübergreifenden Charakter des Wissenschaftlichen Rechnens waren Studenten der Natur- und Ingenieurwissenschaften, der Mathematik und Informatik angesprochen. Die Bewerber

An Innocent-Looking Convergence Problem (2007)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, A Maple Story, A Maple Story, ...

In a recent analysis textbook the following problem is posed (Walter[12], p. 78): Determine all ff such that the sequence x 0 = 1; x 1 = ff; x n+2 = x 2 n+1 \Gamma x n (n 0) (1) is convergent. It...

Analyzing Message Passing Programs on the (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Jim Galarowicz, Jim Galarowicz, Bernd Mohr, Bernd Mohr

Abstract. Writing efficient parallel programs for a massively parallel system like the CRAY T3E is still a difficult task because such programs are typicallyveryly 4 and complex, not trivially...

THE RELATION BETWEEN GALERKIN-TYPE AND 1-NORM QUASI-MINIMAL RESIDUAL ITERATIVE METHODS (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, H. Martin Bcker, H. Martin, B Ucker

Abstract. The main ingredients of any Krylov subspace method for the solution of systems of linear equations with nonsingular, in general non-Hermitian coefficient matrix are the generation of a...

High-level Programming of Massively Parallel Computers based on Shared Virtual Memory (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Michael Gerndt, Michael Gerndt

Highly parallel machines needed to solve compute-intensive scientific applications are based on the distribution of physical memory across the compute nodes. The drawback of such systems is the...

Design Based Methods for the Mapping of Algorithms to Parallel Processors (2007)

Interner Bericht, Heribert C. Burg

Abstract. We present mapping methods which are derived from combinatorics of experimental designs. Experimental designs also known as combinatorial designs are a part of discrete mathematics, which...

e (2007)

Interner Bericht, D. Bardin, D. Lehner, A. Leike

hep-ph/9603438 GENTLE/4fan A package of Fortran programs for the description of e

The SU(2)-Lattice Gauge Theory Simulation Code on the Intel Paragon Supercomputer (2007)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Fritz Gutbrod, Norbert Attig, Michael Weber, ...

We describe the steps which lead to a speed efficiency of about 48 % for a code for the simulation of pure SU(2) Lattice Gauge Theory on the Intel Paragon XP/S Supercomputer. Here the efficiency is...

Flexible and Transparent Fault Tolerance for Distributed Object-Oriented Applications (2007)

Dietmar A. Kottmann, Prof Dr, Alexander B. Schill, Interner Bericht

This report describes an approach enabling automatic structural reconfigurations of distributed applications based on configuration management in order to compensate for node and network failures....

NEC Europe Ltd. Formalizing OpenMP Performance Properties with ASL (2007)

Interner Bericht, Properties Asl, Jesper Larsson Trff, Thomas Fahringer, Thomas Fahringer, Michael Gerndt, ...

Performance analysis is an important step in tuning performance critical applications. It is a cyclic process of measuring and analyzing performance data which is driven by the programmers hypotheses...

John von Neumann-Institut fur Computing Zentralinstitut fur Angewandte Mathematik (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Tobias Oechtering, Tobias Oechtering, Inge Gutheil, Inge Gutheil, ...

The ScaLAPACK [1] library is based on the BLACS (Basic Linear Algebra Communication Subroutines) [2] library. Unfortunately the optimized SHMEM-based CRAY T3E BLACS library contained in libcomm.a has...

Identification and Formalization of Knowledge (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Felix Wolf, Felix Wolf, Bernd Mohr, Bernd Mohr

funded under Contract No. 29488 Acknowledgments We would like to thank all our partners in the APART Working Group for their contribution to this topic. In particular, we would like to express our...

Making C++ Ready for Algorithmic Skeletons (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Jrg Striegnitz, Jorg Striegnitz

Abstract. Many authors have proposed the use of algorithmic skeletons as a high level, machine independent means of developing parallel applications. Since now their implementation and use was...

Unstructured Templates for Programming Irregular Grid Applications on High Performance Computers (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Michael Gerndt, Michael Gerndt

This article presents the concept of unstructured templates developed for SVMFortran, a language for programming scientific applications on massively parallel shared virtual memory systems....

On Riesz and Cesaro Summability of Birkhoff Series (2007)

Ulich Gmbh, Interner Bericht, On Riesz, Cesaro Summability, Horst F. Niemeyer, Horst F. Niemeyer, ...

. For complex discrete Riesz means, as they occur in connection with eigenfunction expansions of Birkhoff type, regularity in the case of positive order, and equivalence to arithmetic means for order...

Preconditioned Solvers for Large Eigenvalue Problems on Massively Parallel Computers and Workstation Clusters (2007)

Te Mathematik, Interner Bericht, Achim Basermann, Bernhard Steffen, A. Basermann, ...

INTRODUCTION The simulation of molecular dynamics and structural mechanics problems is a source of computationally challenging large sparse eigenvalue problems. To solve such problems, we develop...

published as: (2007)

Interner Bericht, Hanns Schulz-mirbach

The Volterra theory of nonlinear systems and algorithms for the construction of invariant image features Hanns Schulz-Mirbach (TUHH)

1 (2007)

Interner Bericht, Hanns Schulz-mirbach, Isaac Weiss

Projective reconstruction from curve

A Learning Algorithm for Structured Invariant Neural Networks (2007)

Interner Bericht, Sabine Kroner

Structured multi-layer feedforward neural networks gain more and more importance in speech and image processing applications. The characteristic of structured neural networks is that a-priori...

Invariant gray scale features (2007)

Interner Bericht, Hanns Schulz-mirbach

Invariant gray scale features Hanns Schulz-Mirbach (TUHH) published as:

1 (2007)

Interner Bericht, Gerald Schreiber

In this paper an algorithm is presented for the parallel implementation and data distribution of structured invariant neural networks (SINN) for invariant pattern recognition. Structured invariant...

y (2007)

Interner Bericht, Peter Rieper, Sabine Kroner, Reinhard Moratz

The capacity or Vapnik-Chervonenkis dimension of a feedforward neural architecture is the maximum number of input patterns that can be mapped correctly to fixed arbitrary outputs. Moreover it is...

Fast adaptive calculation of invariant features (2007)

Interner Bericht, Sabine Kroner, Hanns Schulz-mirbach

The paper presents algorithms for the fast adaptive calculation of invariant features. These are pattern characteristics that remain unchanged under the action of a transformation group. First it is...

--Some Performance Aspects- (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, P. Jansen, M. Marx, W. E. Nagel, M. Vaefien, ...

Today, most of the Cray multiprocessor systems are still used within a multiprogramming environment. In such environments, there are two main issues contributing to whether or not production codes...

Work Package 2 Common Interfaces and Integration Techniques (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Jesper Larsson Trff, Bernd Mohr, Bernd Mohr, Forschungszentrum Julich, ...

Automatic performance tools must of course be tested as to whether they perform their task correctly. Because performance tools are meta-programs, tool testing is more complex than ordinary program...

Towards a Performance Tool Interface for OpenMP: An Approach Based on Directive Rewriting (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Bernd Mohr, Bernd Mohr, Allen D. Malony, Allen D. Malony, ...

Abstract. In this article we propose a "standard " performance tool interface for OpenMP, similar in spirit to the MPI profiling interface in its intent to define a clear and...

Design and Prototype of a Performance Tool Interface for (2007)

Interner Bericht, Bernd Mohr, Bernd Mohr, Allen D. Malony, Allen D. Malony, Sameer Shende, ...

Abstract. This paper proposes a performance tools interface for OpenMP, similar in spirit to the MPI profiling interface in its intent to define a clear and portable API that makes OpenMP execution...

John von Neumann-Institut fur Computing Zentralinstitut fur Angewandte Mathematik (2007)

Forschungszentrum Jlich Gmbh, Interner Bericht, Inge Gutheil, Ruth Zimmermann, I. Gutheil, R. Zimmermann, ...

MPP systems with a message-passing programming paradigm such as CRAY T3E still play an important role in high performance scientific computing and so there is need for basic numerical software for...

Quasi--Minimal Residual Method Based on Coupled Two--Term Recurrences (2007)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, H. Martin Bucker, H. Martin Bucker, Manfred Sauren, ...

Abstract. For the solution of linear systems of equations with unsymmetric coefficient matrix, Freund and Nachtigal (SIAM J. Sci. Comput. 15 (1994), 313--337) proposed a Krylov subspace method called...

1 Modelling of Parallel Processing Tasks by Combinatorial Designs (2007)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Heribert C. Burg, Heribert C. Burg

Combinatorial Designs, also known as experimental designs, historically first were used in agriculture. With the advent of electronic computers together with the significant advancement of...

1 Calculation of Chemical Equilibrium Compositions (2007)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, J Urgen Dornseiffer, Johannes Grotendorst, Johannes Grotendorst, ...

This worksheet will be published in a collection of application on worksheets for the next Maple release.

Forschungszentrum J (2007)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Rudolf Berrendorf, Michael Gerndt, ...

This deliverable gives a detailed language specification of a data-parallel programming language for shared virtual memory systems. In addition to data parallelism, SVM-Fortran supports task...

FORSCHUNGSZENTRUM JLICH GmbH (2007)

Jlich Tel Interner, Forschungszentrum Jülich Gmbh, Interner Bericht, Bart Theelen

Effectively and efficiently implementing parallel programs for computer systems with a large number of processing elements involves the investigation of occurring communication between processes. The...

Approximator Networks and the Principle of Optimal Information Distribution (2007)

Rüdiger W. Brause, Interner Bericht, Dr. Rüdiger W. Brause

It is well known that artificial neural nets can be used as approximators of any continous functions to any desired degree. Nevertheless, for a given application and a given network architecture the...

des John von Neumann-Instituts für Computing (2007)

Jülich Supercomputing Centre, Interner Bericht, Matthias Bolten (hrsg, Eduardo Aguilar, Moreno Bernd Körfgen, Sebastian Birk, ...

Computing (NIC) und hiermit des JSC als wesentlicher Säule des NIC. Um den akademischen Nachwuchs mit verschiedenen Aspekten des Wissenschaftlichen Rechnens vertraut zu machen, führte das JSC in...

des John von Neumann-Instituts für Computing (2007)

Jülich Supercomputing Centre, Interner Bericht, Matthias Bolten (hrsg, Eduardo Aguilar, Moreno Bernd Körfgen, Sebastian Birk, ...

Computing (NIC) und hiermit des JSC als wesentlicher Säule des NIC. Um den akademischen Nachwuchs mit verschiedenen Aspekten des Wissenschaftlichen Rechnens vertraut zu machen, führte das JSC in...

Die Gaststudenten und ihre Betreuer waren: (2006)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Mathias Aust, Guido Arnold, Marcus Richter, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

ii (2006)

Rechnerarchitektur Und Informatik, Neuere Entwicklungen, Interner Bericht

1.1 Der von-Neumann-Rechner.................... 16

ii (2005)

Rechnerarchitektur Und Informatik, Neuere Entwicklungen, Interner Bericht

1.1 Der von-Neumann-Rechner.................... 15

Die Gaststudenten und ihre Betreuer waren: (2005)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Claudia Albrecht, Wolfgang Meyer, Quang Minh, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

Die Gaststudenten und ihre Betreuer waren: (2005)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Claudia Albrecht, Wolfgang Meyer, Quang Minh, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

des John von Neumann-Instituts für Computing (2004)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Nikos Elpidoforou, Paul Gibbon, Ivo Kabadshow, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

ii (2003)

Neuere Entwicklungen, Prof Dr, Hans-jürgen Buhl, Interner Bericht

1.1 Der von-Neumann-Rechner.................... 13

des John von Neumann-Instituts für Computing (2003)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Matthias Bolten, Thomas Müller, Björn Hagemeier, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

des John von Neumann-Instituts für Computing (2003)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Matthias Bolten, Thomas Müller, Björn Hagemeier, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

des John von Neumann-Instituts für Computing (2002)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Katharina Benkert, Bernd Körfgen, Stefan Borowski, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

des John von Neumann-Instituts für Computing (2002)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Katharina Benkert, Bernd Körfgen, Stefan Borowski, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

How to Realize Data-Parallel Algorithmic Skeletons with C++ (2002)

Forschungszentrum Jlich Gmbh, Interner Bericht, Jörg Striegnitz

Many authors have proposed the use of algorithmic skeletons as a high level, machine independent means of developing parallel applications. Until now their implementation and use was restricted to...

Overview Classical Molecular Dynamics and Parallel Computing a (2002)

Forschungszentrum Jülich Gmbh, Interner Bericht, Godehard Sutmann, Godehard Sutmann

Abstract: an overview is given about classical molecular dynamics computer simulation. The main ingredients which enter into a program are presented and special emphasis is put on parallelisation...

Fachbereich Mathematik (7) (2001)

Neuere Entwicklungen, Prof Dr, Hans-jürgen Buhl, Interner Bericht, Praktische Informatik, ...

” Überalterung “ von CISC................ 36 1.3.3 RISC ( = reduced instruction set computer)-Designprinzip 39 1.3.4 Historischer CISC-RISC CPU-Vergleich......... 43 1.4 Der Cache zur...

FORSCHUNGSZENTRUM JLICH GmbH (2001)

Forschungszentrum Jlich Gmbh, Interner Bericht, Inge Gutheil, Inge Gutheil, Forschungszentrum Julich Gmbh

The first installed ATLAS BLAS library on ZAMpano, release 3.0 beta, only contained single processor BLAS for Pentium II processors. It showed good performance and could be used in combination with...

CHASM: Static Analysis and Automatic Code Generation for Improved Fortran 90 and C++ Interoperability (2001)

Forschungszentrum Jülich Gmbh, Interner Bericht, C. E. Rasmussen, K. A. Lindlan, B. Mohr, J. Striegnitz, ...

The relative simplicity and design of the Fortran 77 language allowed for reasonable interoperability with C and C++. Fortran 90, on the other hand, introduces several new and complex features to the...

Automatic Performance Analysis of SMP Cluster Applications (2001)

Forschungszentrum Jülich Gmbh, Interner Bericht, Felix Wolf, Bernd Mohr, Felix Wolf, Bernd Mohr, ...

The expert tool environment provides a complete tracing-based solution for automatic performance analysis of mpi, Openmp, or hybrid applications running on smp cluster machines. expert describes...

Design and Prototype of a Performance Tool Interface for (2001)

Forschungszentrum Jülich Gmbh, Interner Bericht, Bernd Mohr, Allen D. Malony, Sameer Shende, Felix Wolf, ...

Abstract. This paper proposes a performance tools interface for OpenMP, similar in spirit to the MPI profiling interface in its intent to define a clear and portable API that makes OpenMP execution...

Die Gaststudenten und ihre Betreuer waren: (2001)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Ekaterina Borodina, Mathilde Romberg Zam, Thorsten Gellermann, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs mit verschiedenen Aspekten des Wissenschaftlichen Rechnens vertraut zu machen, führte das ZAM in...

Die Gaststudenten und ihre Betreuer waren: (2001)

Forschungszentrum Jülich Gmbh, Interner Bericht, Rüdiger Esser (hrsg, Ekaterina Borodina, Mathilde Romberg Zam, Thorsten Gellermann, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs mit verschiedenen Aspekten des Wissenschaftlichen Rechnens vertraut zu machen, führte das ZAM in...

Towards a Performance Tool Interface for OpenMP: An Approach Based on Directive Rewriting (2001)

Forschungszentrum Jülich Gmbh, Interner Bericht, Bernd Mohr, Allen D. Malony, Sameer Shende, Felix Wolf, ...

Abstract. In this article we propose a “standard” performance tool interface for OpenMP, similar in spirit to the MPI profiling interface in its intent to define a clear and portable API that...

Die Gaststudenten und ihre Betreuer waren: (2001)

Interner Bericht, Beiträge Zum, Wissenschaftlichen Rechnen, Rüdiger Esser (hrsg, Interner Bericht, Beiträge Zum, ...

Computing (NIC) und hiermit des ZAM als wesentlicher Säule des NIC. Um den akademischen Nachwuchs

(letzte Änderung: 14.09.2000) Making C++ Ready for Algorithmic Skeletons (2000)

Forschungszentrum Jülich Gmbh, Interner Bericht, Jörg Striegnitz, Jörg Striegnitz

Abstract. Many authors have proposed the use of algorithmic skeletons as a high level, machine independent means of developing parallel applications. Since now their implementation and use was...

1.1.1 Point toPoint Communication (2000)

Forschungszentrum Jülich Gmbh, Interner Bericht, Tobias Oechtering, Inge Gutheil, Tobias Oechtering, Inge Gutheil, ...

The ScaLAPACK [1] library is based on the BLACS (Basic Linear Algebra Communication Subroutines) [2] library. Unfortunately the optimized SHMEM-based CRAY T3E BLACS library contained in libcomm.a has...

John von Neumann-Institut für Computing (2000)

Forschungszentrum Jülich Gmbh, Interner Bericht, Inge Gutheil, Ruth Zimmermann, I. Gutheil, R. Zimmermann, ...

MPP systems with a message-passing programming paradigm such as CRAY T3E still play an important role in high performance scientific computing and so there is need for basic numerical software for...

Abstract (1999)

Forschungszentrum Jülich Gmbh, Interner Bericht, Thomas Fahringer, Graham Riley, Michael Gerndt, Jesper Larsson Träff

Performance analysis is an important step in tuning performance critical applications. It is a cyclic process of measuring and analyzing performance data which is driven by the programmers hypotheses...

Preface (1999)

Forschungszentrum J, Ulich Gmbh, Interner Bericht

This report is a collection of different applications implemented and running on the massively

Parallel/High-Performance Object-Oriented Scientific Computing (POOSC'99) (1999)

Interner Bericht, Federico Bassetti, Kei Davis, Bernd Mohr (Eds.)

. In this paper we present our object oriented concepts for parallel smoothed particle hydrodynamics simulations based on a 3 year work experience in a government funded project with computer...

Knowledge Specification for Automatic Performance Analysis (1999)

Forschungszentrum Jlich Gmbh, Interner Bericht, Jesper Larsson Träff, Thomas Fahringer, Michael Gerndt, Graham Riley, ...

The lack of a useful and accurate software infrastructure for measuring, modeling, and analyzing the performance of a wide variety of programming paradigms and architecture platforms is a critical...

Higher Order Functions and Partial Application in C++ (1999)

Interner Bericht, Jörg Striegnitz

Polymorphic types, higher order functions, and partial application are common features in functional programming languages. They allow a compact formulation of algorithms and their integration into...

TOPAS - Automatic Performance Statistics Collection on the CRAY T3E (1999)

Interner Bericht, Bernd Mohr, Bernd Mohr, Forschungszentrum Jlich

. TOPAS is a tool to automatically and transparently monitor usage and performance of every parallel job executed on a CRAY T3E. We have modified the UNICOS/mk compiler wrapper scripts to...

Thermalization of Large Fluid Systems consisting of Lennard-Jones Mixtures (1999)

Forschungszentrum Jlich Gmbh, Interner Bericht, Matthias Hloucha

Contents 1 Program MCMD 1 1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....

Identification and Formalization of Knowledge (1999)

Forschungszentrum Jülich Gmbh, Interner Bericht, Jesper Larsson Träff, Thomas Fahringer, Michael Gerndt, Jesper Larsson Träff, ...

We would like to thank all our partners in the APART Working Group for their contribution to this topic. We would like to thank especially those members who participated in the discussions at the work

Performance Issues of Distributed MPI Applications in a German Gigabit Testbed (1999)

Forschungszentrum Jülich Gmbh, Interner Bericht, Thomas Eickermann, Helmut Grund, Jörg Henrichs, Preprint In, ...

Abstract. The Gigabit Testbed West is a testbed for the planned upgrade of the German Scientific Network B-WiN. It is based on a 2.4 Gigabit/second ATM connection between the Research Centre Jülich...

REAPAR User Manual and Reference: Automatic Parallelization of Irregular Recursive Programs (1998)

Stefan U. Hänßgen, Stefan U. H#n#gen, Interner Bericht

This report describes the REAPAR system (REcursive programs Automatically PARallelized) for automatic instrumentation, parallelization strategy selection and parallelization of recursive programs. It...

High-Performance Computer Management Based on Java (1998)

Volker Sander, Te Mathematik, Interner Bericht, Volker S, Volker S, ...

. Coupling of distributed computer resources connectedby a high speed network to one virtual computer is the basic idea of a metacomputer. Access to the metacomputer should be provided by an...

– A Programmable and Extensible Toolkit for Analyzing Event Traces of Message Passing Programs – (1998)

Forschungszentrum Jülich Gmbh, Interner Bericht, Felix Wolf, Bernd Mohr, Felix Wolf, Bernd Mohr, ...

Abstract. This paper describes a new meta-tool named EARL which consists of a new highlevel trace analysis language and its interpreter which allows to easily construct new trace analysis tools....

Einführung in IDL (1998)

Forschungszentrum Jülich Gmbh, Interner Bericht, Marlene Busch, Heinz Heer, Michael Wagener, Marlene Busch Zam, ...

2.2 2. Tag- Techniken und Anwendungen für den IDL Programmierer............ 2.3 3. Tag- Neues mit IDL 5.0 für den erfahrenen IDL Programmierer............ 3 Informationen zu...

Wrappers for Tracing Collective Communication Functions with PAT (1998)

Forschungszentrum Jlich Gmbh, Interner Bericht, Bart Theelen

Effectively and efficiently implementing parallel programs for computer systems with a large number of processing elements involves the investigation of occurring communication between processes. The...

Optimizing and Load Balancing Metacomputing Applications (1998)

Forschungszentrum Jülich Gmbh, Interner Bericht, Jörg Henrichs

While the theoretical possibilities of metacomputing, in the sense of “heterogeneous supercomputing using WANs”, are obvious, it is difficult to utilize this potential in practice. In this paper,...

High-Performance Computer Management Based on Java (1998)

Forschungszentrum Jülich Gmbh, Interner Bericht, Volker S, Dietmar Erwin, Valentina Huber, Volker S, ...

Abstract. Coupling of distributed computer resources connected by a high speed network to one virtual computer is the basic idea of a metacomputer. Access to the metacomputer should be provided by an...

Highly Optimized Code for Lattice Quantum Chromodynamics on the CRAY T3E (1998)

Forschungszentrum Jülich Gmbh, Interner Bericht, Klaus Schilling A, Peer Ueberholz B, Jochen Viehoff B, ...

In lattice quantum chromodynamics, large systems of linear equations have to be solved to compute physical quantities. The availability of efficient parallel Krylov subspace solvers plays a vital...

Fast subsumption checks using anti-links (1997)

Interner Bericht, Fakultat Fur Informatik, Anavai Ramesh, Anavai Ramesh, Neil V. Murray, Neil V. Murray, ...

This report has been submitted for publication elsewhere, and will be copyrighted if accepted.

Delegating Remote Operation Execution in a Mobile Computing Environment (1997)

Dietmar A. Kottmann, Ralph Wittmann, Markus Posur, Interner Bericht

Remote operation execution is nowadays the most popular paradigm used to build distributed systems and applications. This success originates in the simplicity exhibited by programming along the...

Free variable tableaux for propositional modal logics (1997)

Interner Bericht, Fakultat Fur Informatik, Bernhard Beckert, Bernhard Beckert

We present a sound, complete, modular and lean labelled tableau calculus for many propositional modal logics where the labels contain "free" and "universal "...

Partial Order Models for Quantitative Extensions of LOTOS (1997)

Ed Brinksma, Prof Dr. -ing, U. Herzog, Mathematische Maschinen, Joost-pieter Katoen, Interner Bericht, ...

Event structures are a prominent model for non-interleaving concurrency. The use of event structures for providing a compositional non-interleaving semantics to LOTOS without data is studied. In...

A Variant of the Biconjugate Gradient Method Suitable for Massively Parallel Computing (1997)

Forschungszentrum J, Interner Bericht, H. Martin Bücker, H. Martin Bucker, Manfred Sauren, Manfred Sauren

. Starting from a specific implementation of the Lanczos biorthogonalization algorithm, an iterative process for the solution of systems of linear equations with general non-Hermitian coefficient...

New Preconditioned Solvers for Large Sparse Eigenvalue Problems on Massively Parallel Computers (1997)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Achim Basermann, Achim Basermann, Bernhard Steffen, ...

We present preconditioned solvers to find a few eigenvalues and eigenvectors of large dense or sparse symmetric matrices based on the Jacobi-Davidson (JD) method by G. L. G. Sleijpen and H. A. van...

Temporal and Spatial Prewhitening of Multi-Channel MEG Data (1997)

Te Mathematik, Interner Bericht, Heidi Anne Schlitt, Roland Beucker, ...

In this paper we present a prewhitening strategy to decorrelate the noise in space and in time. We discuss in detail how the effects of the filtering can be seen in the singular value decomposition...

A Metacomputer Architecture Based on Cooperative Resource Management (1997)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Volker Sander, Volker Sander

. Computing resources which are transparently available to the user via networked environments are commonly called a metacomputer. To establish a metacomputer there must be a software layer...

Conjugate Gradient and Lanczos Methods for Sparse Matrices on Distributed Memory Multiprocessors (1997)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Achim Basermann, A. Basermann

Conjugate gradient methods to solve sparse systems of linear equations and Lanczos algorithms for sparse symmetric eigenvalue problems play an important role in numerical methods for solving...

Temporal and Spatial Prewhitening of Multi-Channel MEG Data (1997)

Forschungszentrum Jülich Gmbh, Interner Bericht, Heidi Anne Schlitt, R. Beucker, H. A. Schlitt

In this paper we present a prewhitening strategy to decorrelate the noise in space and in time. We discuss in detail how the effects of the filtering can be seen in the singular value decomposition...

High-level Programming of Massively Parallel Computers based on Shared Virtual Memory (1997)

Forschungszentrum Jülich Gmbh, Interner Bericht, Michael Gerndt, Michael Gerndt

Highly parallel machines needed to solve compute-intensive scientific applications are based on the distribution of physical memory across the compute nodes. The drawback of such systems is the...

Particle Simulations on Cray MPP Systems (1997)

Forschungszentrum Jülich Gmbh, Interner Bericht, Christian M. Dury, Renate Knecht, Gerald H. Ristow, Fachbereich Physik, ...

Particle simulations were among the first applications to be implemented on scalar computers over forty years ago, and have since played an important role in many science and engineering...

Unstructured Templates for Programming Irregular Grid Applications on High Performance Computers (1997)

Forschungszentrum Jülich Gmbh, Interner Bericht, Michael Gerndt, Michael Gerndt

This article presents the concept of unstructured templates developed for SVM-Fortran, a language for programming scientific applications on massively parallel shared virtual memory systems....

Object Migration in Non-Monolithic Distributed Applications (1996)

O. Ciupke, D. Kottmann, Interner Bericht, Fakultat Fur Informatik

\Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma \Gamma...

VAMPIR: Visualization and Analysis of MPI Resources (1996)

Forschungszentrum J, Interner Bericht, Wolfgang E. Nagel, Alfred Arnold, Michael Weber, W. E. Nagel, ...

Performance analysis most often is based on the detailed knowledge of program behavior. One option to get this information is tracing. Based on the research tool PARvis, the visualization environment...

Isoefficiency Analysis of Parallel QMR-Like Iterative Methods and its Implications on Parallel Algorithm Design (1996)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, H. Martin Bücker, H. Martin Bucker

A specific problem arising out of electrostatics is taken as an example to demonstrate the process of, firstly, transforming a physical problem into a mathematical model and, secondly, its numerical...

Universitat Karlsruhe (1996)

Fakultat Fur, Interner Bericht, Fakultat Fur Informatik, Bernhard Beckert, Bernhard Beckert

We present a sound, complete, modular and lean labelled tableau calculus for many propositional modal logics where the labels contain "free" and "universal" variables. Our...

A Parallel Version of the Unsymmetric Lanczos Algorithm and its Application to QMR (1996)

Forschungszentrum J, Interner Bericht, H. Martin Bücker, H. Martin Bucker, Manfred Sauren, Manfred Sauren

A new version of the unsymmetric Lanczos algorithm without look-ahead is described combining elements of numerical stability and parallel algorithm design. Firstly, stability is obtained by a coupled...

Programming Shared Virtual Memory Multiprocessors (1995)

Forschungszentrum Jülich Gmbh, Interner Bericht, Michael Gerndt, Michael Gerndt

Highly parallel machines needed to solve compute intensive scientific applications are based on the distribution of physical memory across the compute nodes. The drawback of such systems is the...

leanEA: A Poor Man's Evolving Algebra Compiler, research report (1995)

Interner Bericht, Fakultat Fur Informatik, Bernhard Beckert, Bernhard Beckert, Joachim Posegga, Joachim Posegga

The Prolog program "termexpansion((define C as A with B), (C=?A:-B,!)). termexpansion((transition E if C then D), ((transition E):-C,!,B,A,(transition))):-serialize(D,B,A)....

Dynamic Load Balancing for the Simulation of Granular Materials (1995)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Renate Knecht, Gregory A. Kohring, R. Knecht, ...

We discuss the simulation of granular materials and the problems encountered when parallelizing these simulations. The main problem of inadequately balanced loads can be mitigated via the use of a...

MVP-L Language Report Version 2 (1995)

Interner Bericht, Alfred Bröckers<!--Brockers-->, Alfred Brckers, Christoper M. Lott, Christopher M. Lott, ...

Intellectual control over software development projects requires the existence of an integrated set of explicit models of the products to be developed, the processes used to develop them, the...

A graphical representation schema for the software process modeling language MVP-L (1995)

Interner Bericht, Prof Dr, H. Dieter Rombach, Kaiserslautern Juni, Alfred Bröckers, ...

Experience gathered from applying the software process modeling language MVP---L in software development organizations has shown the need for graphical representations of process models. Project...

Parallelization of the AVL FIRE Benchmark with SVM-Fortran (1995)

Forschungszentrum J, Interner Bericht, Michael Gerndt, Michael Gerndt

This article outlines the parallelization of an irregular grid application with SVMFortran. It describes the different optimizations and their effectiveness. The parallelization was much simplified...

Programming Shared Virtual Memory on the Intel Paragon Supercomputer (1995)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, R. Berrendorf, R. Berrendorf, M. Gerndt, ...

Programming distributed memory systems forces the user to handle the problem of data locality. With message passing the user has not only to map the application to the different processors in a way...

A Remark on Commuting Operator Exponentials (1995)

Ulich Gmbh, Interner Bericht

. In a previous paper the author proved that for square matrices with algebraic entries exp(A)exp(B)=exp(B)exp(A) if and only if AB=BA. This result is extended here to bounded operators on an...

leanEA: A Poor Man's Evolving Algebra Compiler (1995)

Interner Bericht, Fakultat Fur Informatik, Bernhard Beckert, Bernhard Beckert, Joachim Posegga, Joachim Posegga

The Prolog program "termexpansion((define C as A with B), (C=?A:-B,!)). termexpansion((transition E if C then D), ((transition E):-C,!,B,A,(transition ))) :- serialize(D,B,A)....

Object-Oriented Protocol Hierarchies for Distributed Workflow Systems (1995)

P. C. Lockemann, Interner Bericht, Fakultat Fur Informatik

Distributed software systems such as groupware and workflow systems will play a key role in the near future. While numerous models which promise highly sophisticated functionality are proposed in the...

A-Ordered Tableaux (1995)

Interner Bericht, A-ordered Tableaux, Reiner Hähnle, Reiner Hahnle, Stefan Klingenbeck, Stefan Klingenbeck

In resolution proof procedures refinements basedon A-orderings of literals havea long tradition and are well investigated. In tableau proof proceduressuch refinements were only recently introduced by...

Towards an Architecture for WAVE Interpretation in Open Distributed Systems (1995)

Peter M. Borst, Interner Bericht

WAVE is a new model and technology for distributed processing in open systems. Unlike traditional distributed systems it represents a radically novel paradigm, based on program mobility within...

Multi-Object Cooperation in Distributed Object Bases (1995)

D. Kottmann, P. C. Lockemann, Interner Bericht, Fakultat Fur Informatik

It is an emerging trend to build large information systems in a component-based fashion where the components follow the concept of object. Applications are constructed by organizing pre-built objects...

Dynamic Load Balancing for the Simulation of Granular Materials (1995)

Knecht And, Forschungszentrum J, Ulich Gmbh, Interner Bericht, Renate Knecht, Gregory A. Kohring, ...

We discuss the simulation of granular materials and the problems encountered when parallelizing these simulations. The main problem of inadequately balanced loads can be mitigated via the use of a...

Automata-Theoretic Criteria for Feature Interactions in Telecommunications Systems (1995)

Jan Bredereke, Interner Bericht

The feature interaction problem in telecommunications systems increasingly obstructs the evolution of such systems. We develop formal detection criteria which render a necessary (but less than...

leanEA: A Poor Man's Evolving Algebra Compiler, research report (1995)

Interner Bericht, Fakultät Für Informatik, Bernhard Beckert, Bernhard Beckert, Joachim Posegga, Joachim Posegga, ...

“term_expansion((define C as A with B), (C=>A:-B,!)). term_expansion((transition E if C then D), ((transition E):-C,!,B,A,(transition _))):serialize(D,B,A). serialize((E,F),(C,D),(A,B)):-...

Performance Analysis for SVM-Fortran with OPAL (1995)

Forschungszentrum Jülich Gmbh, Interner Bericht, M. Gerndt, A. Krumme, S. Özmen, M. Gerndt, ...

Programming distributed memory parallel computers with message passing is often considered to be a difficult task. To overcome the drawbacks of this programming style, several efforts have been made...

A Case Study on Specification, Detection and Resolution of IN Feature Interactions with Estelle (1994)

Interner Bericht, Jan Bredereke, Reinhard Gotzhein

We present an approach for the treatment of Feature Interactions in Intelligent Networks. The approach is based on the formal description technique Estelle and consists of three steps. For the first...

Intel Paragon XP/S - Architecture, Software Environment, and Performance (1994)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Rüdiger Esser, Rudolf Berrendorf, Rudolf Berrendorf, ...

. The paper describes the hardware and software components of the Intel Paragon XP/S system, a distributed-memory scalable multicomputer. The Paragon processing nodes, which are based on the Intel...

A Verification Concept for SDL Systems and its Application to the Abracadabra Protocol (1994)

Markus Rinderspacher, Interner Bericht

Starting from a purely functional description of a communication protocol, we present a method how correctness proofs including safety- and progress properties can be developed systematically with an...

Automatic Parallelization of a Crystal Growth Simulation Program for Distributed-Memory Systems (1994)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Michael Gerndt, Michael Gerndt

. This article outlines two parallelization tools, i.e. the Vienna Fortran Compilation System and FORGE 90, and discusses their analysis and transformation capabilities in the context of a regular...

A Comparison of QMR, CGS and TFQMR on a Distributed Memory Machine (1994)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Martin Bücker, Martin Bucker, Achim Basermann, ...

For the solution of systems of linear equations with general non-- Hermitian nonsingular coefficient matrices, we propose an implementation of three different algorithms on a parallel machine with...

(Stand 26.04.94) (1994)

Forschungszentrum Jülich Gmbh, Interner Bericht, Dieter Conrads

beschrieben, und es wird kurz dargelegt, welche Entwicklungen in näherer Zukunft im

A Comparison of two Parallelization Strategies for TRACE (1994)

Ulich Gmbh, Interner Bericht, For Trace, Olaf Neuendorf, Michael Gerndt, Michael Gerndt, ...

In this report we compare two different methods of parallelization of a finite element code describing water flow in soils. The first method uses Domain Decomposition based on a parallel Schwarz...

SVM Support in the Vienna Fortran Compilation System (1994)

Interner Bericht, Peter Brezany, Michael Gerndt, Viera Sipkova

Vienna Fortran, a machine-independent language extension to Fortran which allows the user to write programs for distributed-memory systems using global addresses, provides the forall-loop construct...

Performance Analysis Environment for SVM-Fortran Programs (1994)

Forschungszentrum J, Ulich Gmbh, Interner Bericht, Michael Gerndt, M. Gerndt

This report outlines the design of a performance analysis environment for SVMFortran programs. SVM-Fortran is a shared memory parallel programminglanguage developed at KFA for distributed memory...

CM-5, KSR1, Paragon XP/S: a comparative description of massively parallel computers on the basis of a catalog of classifying characteristics (1993)

Tilmann Bönniger, Rüdiger Esser, Ulich Gmbh, Interner Bericht, Dietrich Krekel

In order to facilitate an application-oriented assessment of massively parallel computing systems, a catalog of about 350 classifying characteristics concerning the architecture and software...

SVM Fortran (1993)

Forschungszentrum J, Interner Bericht, Rudolf Berrendorf, Michael Gerndt, Wolfgang E. Nagel, Joachim Pr Ummer, ...

This paper outlines SVM Fortran, an extension of Fortran 77 intended to be used to program shared virtual memory systems or, more general, scalable shared memory systems. The primary target system is...

More Efficient Software Implementations of (Generalized) DES (1990)

Of (generalized Des, Andreas Pfitzmann, Andreas Pfitzmann, Ralf Aßmann, Ralf Amann, ...

By preserving the macro structure of the Data Encryption Standard (DES), but by allowing the user to choose 1. 16.48 independent key bits instead of generating them all using only 56 key bits, 2....

Unconditional Sender and Recipient Untraceability in spite of Active Attacks - Some Remarks (1989)

Michael Waidner, Birgit Pfitzmann, Interner Bericht

...................................................................................................1 2 Unconditional sender...

P Value transfer systems enabling security and unobservability; IFIP/Sec. '86 (1986)

Holger Bürk, Holger Bürk, Andreas Pfitzmann, Andreas Pfitzmann, ...

In present-day cashless payment systems, the banks and (by installing a Trojan Horse) even the manufacturers of the computer equipment used could easily observe who pays what amount to whom and when....

An analysis of transformations (1964)

Prof Dr, Hans Burkhardt, Interner Bericht, Xiaofeng Zhang

An algorithm to group the edge points into digital line segment with Hough transformation is described in this paper. The edge points are mapped onto the parameter domain discretized at specific...