Roller / Benkert / Galle High Performance Computing on Vector Systems 2008

1. Auflage 2008
ISBN: 978-3-540-85869-0
Verlag: Springer-Verlag
Format: PDF
Kopierschutz: Adobe DRM (»Systemvoraussetzungen)

E-Book, Englisch, 242 Seiten

This book covers the results obtained in the Tera op Workbench project during a four years period from 2004 to 2008. The Tera op Workbench project is a colla- ration betweenthe High PerformanceComputingCenter Stuttgart (HLRS) and NEC Deutschland GmbH (NEC-HPCE) to support users to achieve their research goals using high performance computing. The Tera op Workbench supports users of the HLRS systems to enable and - cilitate leading edge scienti c research. This is achieved by optimizing their codes and improving the process work ow which results from the integration of diff- ent modules into a “hybrid vector system”. The assessment and demonstration of industrial relevance is another goal of the cooperation. The Tera op Workbench project consists of numerous individual codes, grouped together by application area and developed and maintained by researchers or c- mercial organizations. Within the project, several of the codes have shown the ab- ity to reach beyond the TFlop/s threshold of sustained performance. This created the possibility for new science and a deeper understanding of the underlying physics. The papers in this book demonstrate the value of the project for different scienti c areas.

Autoren/Hrsg.

Roller, Sabine

Benkert, Katharina

Galle, Martin

Bez, Wolfgang

Kobayashi, Hiroaki

Hirayama, Toshio

Inhaltsverzeichnis

1; I Future Architectures;13
1.1;First Experiences with NEC SX-9;14
1.1.1;Introduction;14
1.1.2;System Overview;16
1.1.3;Performance Evaluation;17
1.1.4;Conclusions;20
1.1.5;References;21
1.2;Scalable Computing in a Hybrid System Architecture;23
1.2.1;Introduction;23
1.2.2;The Need for Scalability;24
1.2.3;Processor Trends;25
1.2.4;The Cray XT5h Hybrid Supercomputer;27
1.2.5;The Cray Cascade Program;29
1.2.6;Conclusion;30
1.2.7;References;30
1.3;Prerequisites for the Productive Usage of Hybrid Systems;32
1.3.1;Introduction;32
1.3.2;HLRS System Overview;33
1.3.3;SW Components for a Hybrid System;34
1.3.3.1;Batch System;35
1.3.3.2;Global Filesystem;36
1.3.3.3;MPI;39
1.3.4;Conclusion;40
1.3.5;References;40
2; II Multiscale and Multiphysics Simulations ;41
2.1;Complex Numerical Simulations in Production Techniques;42
2.1.1;Problem Statement and Motivation;42
2.1.2;Multi-Scale Modeling and Simulation of Manufacturing Systems;43
2.1.2.1;Vision and Scientific Impact;43
2.1.2.2;Challenges, Theoretical Considerations and Definitions;44
2.1.3;A New Concept Towards a Comprehensive and Holistic Factory Simulation Model;45
2.1.3.1;Theoretical Aspects;45
2.1.3.2;Vectorial Modelling and Simulation Approach Based on Business/Organizational Parameters and Technical Tolerances;47
2.1.4;Complex Example in Automotive Coating Industry;49
2.1.5;Conclusions and Future Work;51
2.1.6;References;51
2.2;Multi-scale and Multi-physics Applications --- User Requirements for Future Applications;52
2.2.1;Introduction;52
2.2.2;A Glance at Past, Current and Future Simulation Applications;53
2.2.3;Future Application Perspective;54
2.2.4;Examples;55
2.2.4.1;Dynamic Coating Simulation;55
2.2.4.2;Ocean Water Plant;55
2.2.4.3;Particle-loaden Flows;56
2.2.4.4;Multi-scale Aero-acoustics;57
2.2.5;Future User Perspective;58
2.2.6;Requirements, Architecture and Tools;59
2.2.7;References;59
3; III Grid Computing & Data Analysis;60
3.1;An Application of the NAREGI Grid Middleware to a Nationwide Joint-Use Environment for Computing;61
3.1.1;Introduction;61
3.1.2;NAREGI Grid Middleware;62
3.1.3;Issues;63
3.1.3.1;Resource Provisioning;64
3.1.3.2;Password / Passphrase Management;64
3.1.3.3;Issuance of Grid Certificate;64
3.1.3.4;Grid Operations;65
3.1.4;Application Method;65
3.1.5;Evaluation of Proposed Method;67
3.1.6;Cooperative Evaluation;68
3.1.7;Summary and future works;69
3.1.8;References;69
3.2;Interoperation between Atomic Energy Grid Infrastructure (AEGIS) and Other Grids;71
3.2.1;Introduction;71
3.2.2;History of Grid Computing Technology at CCSE;72
3.2.3;Atomic Energy Grid Infrastructure;74
3.2.4;Interoperability Technology based on Atomic Energy Grid Infrastructure;75
3.2.5;Collaboration with the High Performance Computing Center Stuttgart (HLRS);78
3.2.6;Collaboration with Grid-TLSE Project Partners;80
3.2.7;Collaboration with the U.S. Department of Energy Under the Global Nuclear Energy Partnership;80
3.2.8;Summary;81
3.2.9;References;82
3.3;Parallel File Systems in European Grid Projects;84
3.3.1;Introduction;84
3.3.2;The HLRS Framework;85
3.3.2.1;Target;85
3.3.2.2;Organization;85
3.3.2.3;HLRS Computer Configuration;85
3.3.3;Ex@Grid Framework;86
3.3.3.1;Gauss Center for Supercomputing;86
3.3.3.2;Ex@Grid Design;86
3.3.4;Consolidation in the Datacenter;87
3.3.4.1;Datacenter Ethernet;87
3.3.4.2;IEEE 802.3ar: Congestion Management;87
3.3.5;Multicluster Parallel File Systems;88
3.3.5.1;Multicluster GPFS in DEISA;88
3.3.5.2;Network-centered parallel HSM systems;88
3.3.5.3;Projected mass storage system at HLRS;89
3.3.6;The Teraflop Workbench Concept;89
3.3.6.1;Workflow Example;90
3.3.6.2;Fenfloss;90
3.3.6.3;Demonstration in the HLRS Cave;90
3.3.6.4;Virtual Tour of Kiebingen Water Power Plant;91
3.3.7;Conclusion;91
3.3.8;References;92
3.4;Development of Cognitive Methodology based Data Analysis System;93
3.4.1;Introduction;93
3.4.2;The Basic Idea of CDAS;95
3.4.2.1;Data Analysis Process;95
3.4.2.2;VV and DD Functions;95
3.4.2.3;Synthesis Function;96
3.4.2.4;Evaluation and Judgment;96
3.4.3;System Configuration;97
3.4.3.1;Flow of Data Analysis by CDAS;97
3.4.3.2;Implementation on a Grid Computing;98
3.4.4;Result and Discussion;98
3.4.5;Conclusions and Future Works;100
3.4.6;References;101
4; IV Chemical Applications;102
4.1;3D-Flame Modelling in Power Plant Applications;103
4.1.1;Introduction;103
4.1.2;Flame modelling tool and computer hardware;104
4.1.3;Neurath A/B boiler model;106
4.1.4;Simulation results;107
4.1.5;Summary;112
4.1.6;References;112
4.2;Hierarchical Modeling of Combustion Processes;113
4.2.1;Introduction;113
4.2.2;Dynamics of Reacting Flows;114
4.2.3;Detailed Models;116
4.2.3.1;Chemical kinetics;117
4.2.3.2;Chemistry-Turbulence Coupling;117
4.2.3.3;Modeling of Multi-Phase Processes;118
4.2.4;Model Reduction;120
4.2.4.1;Chemical Kinetics;121
4.2.4.2;Chemistry-Turbulence Interaction;123
4.2.4.3;Multi-Phase Processes;125
4.2.5;Coupling of the Sub-Models;125
4.2.6;Summary;127
4.2.7;References;128
4.3;Understanding Molecular Recognition and Self-Assembly from Large-Scale Numerical Simulations;130
4.3.1;Introduction;130
4.3.2;Computational Method;132
4.3.3;Results and Discussion;132
4.3.4;Conclusions;137
4.3.5;References;138
4.4;Large Scale Particle-in-cell Plasma Simulation;139
4.4.1;Introduction;139
4.4.2;Parallelization of 3D Particle-in-cell Code using High Performance Fortran;140
4.4.3;PIC Simulation of Blob Transport;141
4.4.4;Conclusion;143
4.4.5;References;144
4.5;Multi-scale Modeling of Crack Propagation;145
4.5.1;Introduction;145
4.5.2;Multi-scale Model of SCC;146
4.5.2.1;Oxygen Embrittlement Mechanism;146
4.5.2.2;Mechanics Modeling of Crack Growth;147
4.5.2.3;Oxygen Diffusion Modeling;149
4.5.3;Simulation Details and Results;149
4.5.4;Conclusions;151
4.5.5;References;152
5; V Climate Modeling, Hydro- and Aerodynamics;153
5.1;The Climate Model ECHAM5 on NEC SX-8;154
5.1.1;Introduction;154
5.1.2;Runtime Analysis;155
5.1.3;Optimizations;156
5.1.3.1;Single CPU Optimizations;156
5.1.3.2;Scalability Optimizations;157
5.1.4;Performance Results;157
5.1.4.1;Single CPU Performance;157
5.1.4.2;Scalability;158
5.1.5;Conclusions;160
5.1.6;References;161
5.2;A Large Spectrum of Free Oceanic Oscillations;162
5.2.1;Introduction;162
5.2.2;State of Knowledge;163
5.2.3;Free Oceanic Oscillation Model with Consideration of LSA;164
5.2.3.1;Theory;165
5.2.3.2;The Implicitly Restarted Arnoldi Method;166
5.2.3.3;The Parallelization with MPI;168
5.2.3.4;The Performance of the Model;168
5.2.4;Results;169
5.2.4.1;Gravitational Modes;170
5.2.4.2;Vorticity Modes;170
5.2.5;Conclusion;173
5.2.6;References;173
5.3;Direct Numerical Simulation of Controlled Shear Flows;175
5.3.1;Laminar-Flow-Control Case;175
5.3.1.1;Introduction;175
5.3.1.2;Numerical Model;177
5.3.1.3;Secondary Instability and Control Setup;178
5.3.1.4;Control Results and Conclusions;179
5.3.2;Effusion Cooling Case at