High Performance Computing on Vector Systems 2008

This book presents the state of the art in high-performance computing and simulation on modern supercomputer architectures. It covers trends in hardware and software development in general and specifically the future of vector-based systems and heterogeneous architectures. The application contributions cover computational fluid dynamics, fluid-structure interaction, physics, chemistry, astrophysics, and climate research. Innovative fields like coupled multi-physics or multi-scale simulations are presented. All papers were chosen from presentations given at the seventh Teraflop Workshop, held in November 2007 at Tohoku University, Japan; the eighth Teraflop Workshop, held in April 2008 at the Höchstleistungsrechenzentrum Stuttgart (HLRS), Germany; and at the Japan Atomic Energy Agency in April 2008. Front Matter....Pages i-xiii Front Matter....Pages 1-1 First Experiences with NEC SX-9....Pages 3-11 Scalable Computing in a Hybrid System Architecture....Pages 13-21 Prerequisites for the Productive Usage of Hybrid Systems....Pages 23-31 Front Matter....Pages 33-33 Complex Numerical Simulations in Production Techniques....Pages 35-44 Multi-scale and Multi-physics Applications — User Requirements for Future Applications....Pages 45-52 Front Matter....Pages 53-53 An Application of the NAREGI Grid Middleware to a Nationwide Joint-Use Environment for Computing....Pages 55-64 Interoperation between Atomic Energy Grid Infrastructure (AEGIS) and Other Grids....Pages 65-77 Parallel File Systems in European Grid Projects....Pages 79-87 Development of Cognitive Methodology based Data Analysis System....Pages 89-97 Front Matter....Pages 99-99 3D-Flame Modelling in Power Plant Applications....Pages 101-110 Hierarchical Modeling of Combustion Processes....Pages 111-127 Understanding Molecular Recognition and Self-Assembly from Large-Scale Numerical Simulations....Pages 129-137 Large Scale Particle-in-cell Plasma Simulation....Pages 139-144 Multi-scale Modeling of Crack Propagation....Pages 145-152 Front Matter....Pages 193-193 The Climate Model ECHAM5 on NEC SX-8....Pages 155-162 A Large Spectrum of Free Oceanic Oscillations....Pages 163-175 Direct Numerical Simulation of Controlled Shear Flows....Pages 177-193 Fluid-Structure Interaction in Turbine Simulation....Pages 195-213 Heterogeneous Parallel Aero-Acoustics Using PACX-MPI....Pages 215-222 Meandering of Wing-Tip Vortices Interacting with a Cold Jet in the Extended Wake....Pages 223-242 First experiences with NEC SX-9 -- Scalable computing in a hybrid system architecture -- Prerequisites for the productive usage of hybrid systems -- Complex numerical simulations in production techniques -- Multi-scale and multi-physics applications : user requirements for future applications -- Application of the NAREGI grid middleware to a nationwide joint-use environment for computing -- Interoperation between Atomic Energy Grid Infrastructure (AEGIS) and other grids -- Parallel file systems in European grid projects -- Development of cognitive methodology based data analysis system -- 3D-flame modelling in power plant applications -- Hierarchical modeling of combustion processes -- Understanding molecular recognition and self-assembly from large-scale numerical simulations -- Large scale particle-in-cell plasma simulation -- Multi-scale modeling of crack propagation -- Climate model ECHAMS on NEC SX-8 -- Large spectrum of free oceanic oscillations -- Direct numerical simulation of controlled shear flows -- Fluid-structure interaction in turbine simulation -- Heterogeneous parallel aero-acoustics using PACX-MPI -- Meandering of wing-tip vortices interacting with a cold jet in the extended wake 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. Presents the developments in high-performance computing and simulation on modern supercomputer architectures. This book covers trends in hardware and software development in general and specifically the vector-based systems and heterogeneous architectures. It presents innovative fields like coupled multi-physics or multi-scale simulations