Parallel Computational Fluid Dynamics '99 (eBook)

Towards Teraflops, Optimization and Novel Formulations

A. Ecer, P. Fox, D. Keyes, Jacques Periaux, N. Satofuka (Autoren)

eBook Download: PDF

2000 | 1. Auflage
478 Seiten
Elsevier Science (Verlag)
978-0-08-053838-9 (ISBN)

Infos

Contributed presentations were given by over 50 researchers representing the state of parallel CFD art and architecture from Asia, Europe, and North America. Major developments at the 1999 meeting were: (1) the effective use of as many as 2048 processors in implicit computations in CFD, (2) the acceptance that parallelism is now the 'easy part' of large-scale CFD compared to the difficulty of getting good per-node performance on the latest fast-clocked commodity processors with cache-based memory systems, (3) favorable prospects for Lattice-Boltzmann computations in CFD (especially for problems that Eulerian and even Lagrangian techniques do not handle well, such as two-phase flows and flows with exceedingly multiple-connected demains with a lot of holes in them, but even for conventional flows already handled well with the continuum-based approaches of PDEs), and (4) the nascent integration of optimization and very large-scale CFD. Further details of Parallel CFD'99, as well as other conferences in this series, are available at http://www.parcfd.org

Inhalt

Front Cover 1
Parallel Computational Fluid Dynamics: Towards Teraflops, Optimization and Novel Formulations 4
Copyright Page 5
TABLE OF CONTENTS 16
Preface 6
Acknowledgments 8
List of Scientific Committee Members 10
List of Participants 11
Conference Photograph 14
PART I: PLENARY PAPERS 22
Chapter 1. Parallel Multigrid Solution and Optimization in Compressible Flow Simulation and Design 24
Chapter 2. High-Performance Spectral Element Algorithms and Implementations 38
Chapter 3. Operator Splitting and Domain Decomposition for Multiclusters 48
Chapter 4. Development of the Earth Simulator 58
Chapter 5. Virtual Manufacturing and Design in the Real World — Implementation and Scalability on HPCC Systems 68
Chapter 6. Large-Scale Parallel Viscous Flow Computations Using an Unstructured Multigrid Algorithm 78
PART II: CONTRIBUTED PAPERS 94
Chapter 7. Efficient Parallel Implementation of a Compact Higher-Order Maxwell Solver Using Spatial Filtering 96
Chapter 8. Parallel Two-Level Methods for Three-dimensional Transonic Compressible Flow Simulations on Unstructured Meshes 110
Chapter 9. Parallel Calculation of Helicopter BVI Noise by a Moving Overlapped Grid Method 118
Chapter 10. Domain Decomposition Implementations for Parallel Solutions of 3D Navier- Stokes Equations 126
Chapter 11. Parallel Implementation of the Discontinuous Galerkin Method 136
Chapter 12. Numerical Simulations of Complex Flows with Lattice-Boltzmann Automata on Parallel Computers 144
Chapter 13. Parallel Preconditioners for KKT Systems Arising in Optimal Control of Viscous Incompressible Flows 152
Chapter 14. Parallel Polynomial Preconditioners for the Analysis of Chaotic Flows in Rayleigh-Benard Convection 160
Chapter 15. An Object-Oriented Software Framework for Building Parallel Navier-Stokes Solvers 168
Chapter 16. Parallel NAS3D: An Efficient Algorithm for Engineering Spray Simulations using LES 176
Chapter 17. Dynamic Load Balancing for Parallel CFD on NT Networks 186
Chapter 18. Parallel Computer Simulation of a Chip Bonding to a Printed Circuit Board: In the Analysis Phase of a Design and Optimization Process for Electronic Packaging 194
Chapter 19. Implicit-explicit Hybrid Schemes for Radiation Hydrodynamics Suitable for Distributed Computer Systems 200
Chapter 20. Computations of Three-dimensional Compressible Rayleigh-Taylor Instability on SGI/Cray T3E 210
Chapter 21. An Algorithm for Reducing Communication Cost in Parallel Computing 220
Chapter 22. Domain Decomposition Solution of Incompressible Flows using Unstructured Grids with pP2P1 Elements 228
Chapter 23. Parallel Agglomeration Strategies for Industrial Unstructured Solvers 236
Chapter 24. Parallel Computation of Turbulent Flows Around Complex Geometries on Hybrid Grids with the DLR-TAU Code 244
Chapter 25. Parallel Implementation of a Commonly Used Internal Combustion Engine Code 252
Chapter 26. Towards Realistic Performance Bounds for Implicit CFD Codes 262
Chapter 27. A Parallel Computing Framework for Dynamic Power Balancing in Adaptive Mesh Refinement Applications 270
Chapter 28. A Two-Level Aggregation-Based Newton-Krylov-Schwarz Method for Hydrology 278
Chapter 29. The Next Four Orders of Magnitude in Performance for Parallel CFD 286
Chapter 30. Design of Large-Scale Parallel Simulations 294
Chapter 31. An Efficient Storage Technique for Parallel Schur Complement Method and Applications on Different Platforms 302
Chapter 32. Applications of the Smoothed Particle Hydrodynamics method: The Need for Supercomputing 310
Chapter 33. Parallel Multigrid with Blockwise Smoothers for Multiblock Grids 318
Chapter 34. An Artificial Compressibility Solver for Parallel Simulation of Incompressible Two-Phase Flows 326
Chapter 35. Implementation of a Parallel Framework for Aerodynamic Design Optimization on Unstructured Meshes 334
Chapter 36. Validation of a Parallel Version of the Explicit Projection Method for Turbulent Viscous Flows 342
Chapter 37. Parallel Solution of Multibody Store Separation Problems by a Fictitious Domain Method 350
Chapter 38. Efficient Parallel-by-Line Methods in CFD 358
Chapter 39. An Automatable Generic Strategy for Dynamic Load Balancing in a Parallel Structured Mesh CFD Code 366
Chapter 40. Parallel Performance of a Zonal Navier-Stokes Code on a Missile Flowfield 376
Chapter 41. Parallel Computation of Three-dimensional Two-phase Flows By Lattice- Boltzmann Method 384
Chapter 42. Parallelization and Optimization of a Large Eddy Simulation Code using and OpenMP for SGI Origin2000 Performance 392
Chapter 43. Calculation of Unsteady Incompressible Flows on a Massively Parallel Computer Using the B.F.C. Coupled Method 402
Chapter 44. A Parallel Algorithm for the Detailed Numerical Simulation of Reactive Flows 410
Chapter 45. Parallelization of the Edge-based Stabilized Finite Element Method 418
Chapter 46. The Need for Multigrid for Large Computations 428
Chapter 47. Parallel Computations of Unsteady Euler Equations on Dynamically Deforming Unstructured Grids 436
Chapter 48. Parallelization and MPI Performance of Thermal Lattice Boltzmann Codes for Fluid Turbulence 444
Chapter 49. Parallel Computation of Two-Phase Flows Using the Immiscible Lattice Gas 452
Chapter 50. Parallel Performance Modeling of an Implicit Advection-Diffusion Solver 460
Chapter 51. A Parallel 3D Fully Implicit Unsteady Multiblock CFD Code Implemented on a Beowulf Cluster 468

Produkt-Details

• Erscheint lt. Verlag: 18.10.2000
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Informatik ► Theorie / Studium
• Themenwelt: Mathematik / Informatik ► Mathematik ► Analysis
• Themenwelt: Naturwissenschaften ► Physik / Astronomie ► Strömungsmechanik
• Themenwelt: Technik
• ISBN-10: 0-08-053838-X / 008053838X
• ISBN-13: 978-0-08-053838-9 / 9780080538389