Applied Parallel Computing: Advanced Scientific Computing

This book constitutes the refereed proceedings of the 6th International Conference on Applied Parallel Computing, PARA 2002, held in Espoo, Finland, in June 2002.The 50 revised full papers presented together with nine keynote lectures were carefully reviewed and selected for inclusion in the proceedings. The papers are organized in topical sections on data mining and knowledge discovery, parallel program development, practical experience in parallel computing, computer science, numerical algorithms with hierarchical memory optimization, numerical methods and algorithms, cluster computing, grid and network technologies, and physics and applications.

Inhalt

Keynote Lectures.- Enabling Numerical and Software Technologies for Studying the Electrical Activity in Human Heart.- Parallel Patient-Specific Computational Haemodynamics.- High Performance Computing, Computational Grid, and Numerical Libraries.- Grid Computing: Enabling a Vision for Collaborative Research.- HPC - What Might the Future Hold?.- Multi-physics and Multi-scale Modelling of Materials Processing.- Co-array Fortran for Full and Sparse Matrices.- Measuring the Local Geometry of Valleys in Complex Energy Landscapes by Exhaustive Exploration: The Lid Method.- An Overview of an Architecture Proposal for a High Energy Physics Grid.- Datamining and Knowledge Discovery.- A Data Mining Architecture for Clustered Environments.- Automated Fitting and Rational Modeling Algorithm for EM-Based S-Parameter Data.- A Proposal of High Performance Data Mining System.- A Quasi-Parallel Realization of the Investment Frontier in Computer Resource Allocation Using Simple Genetic Algorithm on a Single Computer.- Parallelism in Knowledge Discovery Techniques.- Parallel Program Development.- A New Approach to Parallel Debugger Architecture.- ALCOR - An Algorithmic Concept Recognition Tool to Support High Level Parallel Program Development.- MPIT - Communication/Computation Paradigm for Networks of SMP Workstations.- Code Optimization Techniques of Data-Intensive Tasks onto Statically Scheduled Architectures: Optimal Performance on the TigerSharc.- Practical Experiences in Parallel Computing.- PIT: A Library for the Parallelization of Irregular Problems.- Parallel Information Retrieval with Query Expansion.- Reducing Communication Cost for Parallelizing Irregular Scientific Codes.- Implementation of Parallel Collection Equi-Join Using MPI.- Practical Experiences in Parallelizing Existent Computer Programs.- Computer Science.- On the Evaluation of the Distributed Objects and Mobile Agents Programming Models for a Distributed Optimization Application.- A Parallel Transitive Closure Computation Algorithm for VLSI Test Generation.- Space-Efficient First Race Detection in Shared Memory Programs with Nested Parallelism.- A Practical Method for On-the-Fly Data Race Detection.- Parallelisms in MPEG and Its Applications to 3-D Visualization.- Numerical Algorithms with Hierarchical Memory Optimization.- A Recursive Formulation of the Inversion of Symmetric Positive Definite Matrices in Packed Storage Data Format.- Parallel Two-Sided Sylvester-Type Matrix Equation Solvers for SMP Systems Using Recursive Blocking.- Performance Optimization of 3D Multigrid on Hierarchical Memory Architectures.- Numerical Methods and Algorithms A.- Parallel and Blocked Algorithms for Reduction of a Regular Matrix Pair to Hessenberg-Triangular and Generalized Schur Forms.- Enhanced Services for Remote Model Reduction of Large-Scale Dense Linear Systems.- HUTI: Framework for Iterative Solvers.- Numerical Methods and Algorithms B.- A Block Fourier Decomposition Method.- New Parallel Architecture for Modular Multiplication and Squaring Based on Cellular Automata.- A Parallel Implementation of the Tree-Structured Self-Organizing Map.- A Blocking Algorithm for Parallel 1-D FFT on Shared-Memory Parallel Computers.- Numerical Methods and Algorithms C.- A Technique for Parallel Loop Execution.- A Self-Adaptable Distributed Evolutionary Algorithm to Tackle Space Planning Problems.- Efficient Parallel Solution to Calculate All Cycles in Graphs.- Experiences with Cluster Computing A.- Scheduling Strategies for Master-Slave Tasking on Heterogeneous Processor Grids.- High-Performance Computing: Past, Present, and Future.- Fast MPI Broadcasts through Reliable Multicasting.- A Framework for Building Distributed Data Flow Chains in Clusters.- Experiences with Cluster Computing B.- Performance of an IBM Pwr4 Node for the GEMS TD Codes and Parallacs.- A Cluster-Based Solution for a High Performance Air Quality Simulation.- Compiler-Controlled Parallelism-Independent Scheduling for Parallel and Distributed Systems.- Optimization of Parallel Algorithms on Cluster of SMP's.- XII Grid and Network Technologies.- Reliability Bounds for Large Multistage Interconnection Networks.- Grid Technology with Dynamic Load Balancing for Monte Carlo Simulations.- A Parallel Grid Based PSE for EHL Problems.- A "Single-Box" Re-routing Architecture for a 3-Stage Rearrangeable CLOS Interconnection Networks.- Enhancing Load Balancing in a Data-Parallel GSM Network Simulation through Application-Specific Information.- XIII Physics and Applications.- Automated Tracking of 3-D Overturn Patches in Direct Numerical Simulation of Stratified Homogeneous Turbulence.- Improving Load Balance in a Weather Code: Asynchronous Output in HIRLAM with MPI.- Parallel Simulation of Photorefractive Material for the Design of All-Optical Components.- Scalable Sparse Matrix Techniques for Modeling Crack Growth.- Parallelization of a Lattice Boltzmann Suspension Flow Solver.