Parallel Virtual Machine - EuroPVM'96 - Third European PVM Conference, Munich, Germany, October 7-9, 1996. Proceedings

This book constitutes the refereed proceedings of the Third European Conference on the Parallel Virtual Machine, EuroPVM '96, the 1996 European PVM users' group meeting, held in Munich, Germany, in October 1996.
The parallel virtual machine, PVM, was developed at the University of Tennessee and Oak Ridge National Laboratory in cooperation with Emory University and Carnegie Mellon University to support distributed computing. This volume comprises 51 revised full contributions devoted to PVM. The papers are organized in topical sections on evaluation of PVM; Applications: CFD solvers; tools for PVM; non-numerical applications; extensions to PVM; etc.

List of contents

Advanced programming in PVM.- Tools for monitoring, debugging, and programming in PVM.- The status of the MPI Message-Passing standard and its relation to PVM.- A comparison of message-passing parallelization to shared-memory parallelization.- Practical experiments to improve PVM algorithms.- PVM, computational geometry, and parallel computing course.- Parallelization of a state-of-the-art industrial CFD package for execution on networks of workstations and massively parallel processors.- A parallel multigrid method for the prediction of incompressible flows on workstation clusters.- An abstract data type for parallel simulations based on sparse grids.- The PVM system: Status, trends, and directions.- OCM - An OMIS compliant monitoring system.- State based visualization of PVM applications.- Workstation cluster as a parallel hardware environment for high performance computing.- Using PVM to implement PPARDB/PVM, a portable parallel database management system.- SPTHEO - A PVM-based parallel theorem prover.- Parallel construction of finite solvable groups.- Proving properties of PVM applications - A case study with CoCheck.- Molecular dynamics simulations on Cray clusters using the SCIDDLE-PVM environment.- Parallelization of the SPAI preconditioner in a master-slave configuration.- Parallel lanczos algorithm on a CRAY-T3D combining PVM and SHMEM routines.- PVM-implementation of sparse approximate inverse preconditioned for solving large sparse linear equations.- Experience with PVM in an industrial environment.- Taskers and General Resource Managers: PVM supporting DCE Process Management.- A PVM implementation of a portable parallel image processing library.- Porting of an empirical tight-binding Molecular Dynamics code on MIMD platforms.- Automatic test patterngeneration with optimal load balancing.- Parallel crystal growth for chip placement.- Efficient message passing on shared memory multiprocessors.- Development of PVM code for a Low Latency Switch Based Interconnect.- PSPVM: Implementing PVM on a high-speed interconnect for workstation clusters.- Implementing the ALWAN communication and data distribution library using PVM.- Transformation of functional programs into data flow graphs implemented with PVM.- Computer-assisted generation of PVM/C++ programs using CAP.- Load balancing for computation of steady viscous flows using adaptive unstructured grids.- Using the ALDY load distribution system for PVM applications.- Parallel Virtual Machine for Windows95.- Using PVM in wireless network environments.- Parallel image compression on a workstationcluster using PVM.- Genetic selection and generation of textural features with PVM.- {0,1}-Solutions of integer linear equation systems.- Extending synchronization PVM mechanisms.- PLS: A parallel linear solvers library for Domain Decomposition methods.- Parallel integration across time of initial value problems using PVM.- Distributed computation of ribbon tableaux and spin polynomials.- Parallel model-based diagnosis using PVM.- Running PVM-GRACE on workstation clusters.- Search of molecular ground state via genetic algorithm; Implementation on a hybrid SIMD-MIMD platform.- Using PVM in the simulation of a hybrid dataflow architecture.- Managing nondeterminism in PVM programs.- Parallelization of an evolutionary neural network optimizer based on PVM.- On the synergistic use of parallel exact solvers and heuristic/stochastic methods for combinatorial optimisation problems: A study case for the TSP on the Meiko CS-2.

About the author

Prof. Dr. Arndt Bode wurde 1987 auf den Lehrstuhl für Rechnertechnik und Rechnerorganisation in der Fakultät für Informatik der TUM berufen. Seine wichtigsten Arbeitsgebiete sind technische Informatik, Rechnerarchitektur, Supercomputer, parallele und verteilte Systeme, Entwicklungswerkzeuge sowie verteilte Anwendungen. Daneben gilt sein besonderes Engagement der fachübergreifenden Kooperation in Forschung und Lehre für Informatikprodukte, der Standardisierung von Grundsoftware für parallele und verteilte Systeme sowie dem Technologietransfer.§Von Bodes hoher wissenschaftlicher Qualifikation zeugen über 200 Fachpublikationen, darunter Bücher über Rechnerarchitektur, Mikroprogrammierung, Mikroprozessoren und Anwendungen. Diese wurde unter anderem honoriert mit dem "Fellowship des Department of Employment, Education and Training", Canberra, Australien, dem "Best University Research Demonstration Award CONPAR 1990", Zürich und dem "Award for Online Visualization, CONPAR 90", ETH Zürich. Bode ist auch Herausgeber verschiedener nationaler und internationaler Buchreihen und wissenschaftlicher Zeitschriften, seit 1999 Hauptherausgeber des Informatik-Spektrum, Springer Verlag. §Bode ist national und international ein sehr gefragter Gutachter und Mitglied in zahlreichen Fachgesellschaften. Seit 1994 gehört er der Kommission für Informatik der Bayerischen Akademie der Wissenschaften an, seit 1999 der Kommission für Rechenanlagen der Deutschen Forschungsgemeinschaft. Ebenso übt er zahlreiche Tätigkeiten als Berater für nationale und internationale Unternehmen aus. §Im Februar 2005 ist Prof. Bode für weitere drei Jahre in seinem Amt als Vizepräsident und CIO der TU München bestätigt worden.

Product details

• Assisted by: Arndt Bode (Editor), Jac Dongarra (Editor), Jack Dongarra (Editor), Thomas Ludwig (Editor), Thomas Ludwig et al (Editor), Vaidy Sunderam (Editor)
• Publisher: Springer, Berlin
• Languages: English
• Product format: Paperback / Softback
• Released: 15.02.2001
• EAN: 9783540617792
• ISBN: 978-3-540-61779-2
• No. of pages: 373
• Weight: 576 g
• Illustrations: XV, 373 p.
• Series: Lecture Notes in Computer Science; Lecture Notes in Computer Science
• Subject: Natural sciences, medicine, IT, technology > IT, data processing > IT