Parallel Computational Fluid Dynamics 2008

Name: Parallel Computational Fluid Dynamics 2008 | Parallel Numerical Methods, Software Development and Applications
Brand: Springer
Price: 149.79 EUR
Availability: OnlineOnly

Parallel Numerical Methods, Software Development and Applications

Damien Tromeur-Dervout Gunther Brenner David R. Emerson Jocelyne Erhel(Editor)

Springer (Publisher)

1st Edition

Published on 21. September 2010

XIV, 438 pages

E-Book

PDF with digital watermarking

System requirements

978-3-642-14438-7 (ISBN)

€149.79incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Parallel Computational Fluid Dynamics 2008
Preface
Contents
Part I Invited speakers
Large Scale Computations in Nuclear Engineering: CFD for Multiphase Flows and DNS for Turbulent Flows with/without Magnetic Field
1 Numerical Simulation of Boiling Phenomena
1.1 Numerical Simulation based on MARS
1.2 Results and Discussions on Forced Convective Flow Boiling
2 CFD for Two-Phase Flow Behaviors in Nuclear Reactor
2.1 Numerical Simulation of Two-Phase Flow Behavior in 37 RMWR Fuel Rods
2.2 Results and Discussions
3 DNS for Turbulent Flows with/without Magnetic Field
4 Conclusion
Scalable algebraic multilevel preconditioners with application to CFD
1 Introduction
2 Algebraic Multilevel Schwarz Preconditioners
3 The MLD2P4 software package
4 Using MLD2P4 in the LES of turbulent channel flows
5 Numerical experiments
6 Conclusions
Acceleration of iterative solution of series of systems due to better initial guess
1 Introduction
2 Linear systems with the same matrix
3 Linear systems with different matrices
4 Series of nonlinear systems
Part II Optimisation in Aerodynamics Design
Aerodynamic Study of Vertical Axis Wind Turbines
1 Introduction
2 Computational models
2.1 Aerodynamic parameter model
2.2 CFD Analysis
3 Results
4 Conclusion
Parallel Shape Optimization of a Missile on a Grid Infrastructure
1 Introduction
2 Problem Definition
2.1 The Cruise Flight
2.2 Shape Optimization
3 Method
3.1 Genetic Algorithms
3.2 Shape Optimization of a Missile
3.3 Implementation of Problem on GridAE
4 Test Case
5 Conclusions
Analysis of Aerodynamic Indices for Racing Sailing Yachts: a Computational Study and Benchmark on up to 128 CPUs.
1 Introduction
2 Experimental measurements
3 Numerical analysis
4 Results
5 Conclusion
Parallel implementation of fictitious surfaces method for aerodynamic shape optimization
1 INTRODUCTION
2 OPTIMIZATION METHOD
2.1 Design Space and Search Algorithm
2.2 Computational Efficiency
2.3 Incorporation of fictitious surfaces
3 ANALYSIS OF RESULTS
4 CONCLUSIONS
Path Optimization of Dual Airfoils Flapping in a Biplane Configuration with RSM in a Parallel Computing Environment
1 INTRODUCTION
2 Response Surface Methodology, RSM
3 Periodic Path defined by NURBS
4 Numerical Method
4.1 Flow Solver
4.2 Flapping Motion
4.3 Optimization based on the Steepest Ascent Method
4.4 Parallel Processing
5 Results and Discussion
5.1 Validation Study
5.2 Optimization for Dual Airfoils
6 Conclusion
Part III Grid methods
Convergence Improvement Method for Computational Fluid Dynamics Using Building-Cube Method
1 Introduction
2 Numerical Methods
2.1 Computational Gird of BCM
2.2 Flow Simulation
2.3 Parallelization
2.4 Validation
2.5 Implicit Residual Smoothing (IRS)
3 Results and Discussion
4 Conclusion
Aerodynamic Analysis of Rotor Blades using Overset Grid with Parallel Computation
1 Introduction
2 Validation of Parallelized Solver
2.1 DLR-F6
3 Rotor Blades
3.1 Overset Grids
3.2 Numerical Analysis Method
3.3 Overset Grid Technique and Parallel Implementation
4 Rotor Flight Test and Results
4.1 Hovering Flight Test
4.2 Non-lift Forwarding Flight Test
4.3 Forwarding Flight Test
5 Conclusion
Large scale massively parallel computations with the block-structured elsA CFD software
1 Introduction
2 Detailed analysis of elsA parallel performance
2.1 Load balancing
2.2 MPI message scheduling
2.3 Test of different implementations of MPI messages
Use of MPI_Sendrecv_replace()
Concatenation of messages to reduce latency
Use of MPI_Bsend()
Replacing blocking with non-blocking MPI routines
2.4 Influence of block splitting on numerical convergence
3 elsA parallel efficiency
3.1 Test case description
3.2 IBM BlueGene/L
3.3 BULL Novascale (Itanium Montecito 1.6GHz)
3.4 SGI Altix XE (Xeon Woodcrest 3GHz dual core)
3.5 HP Cluster (Opteron AMD 2.4 GHz)
Applications on Hybrid Unstructured Moving Grid Method for Three-Dimensional Compressible Flows
1 Introduction
2 Hybrid Unstructured Moving-Grid Finite Volume Method
2.1 Governing Equation
2.2 Numerical Algorithm
3 Application
3.1 Calculating condition
3.2 Result
4 Conclusions
Progressive Development of Moving-Grid Finite-Volume Method for Three-Dimensional Incompressible Flows
1 Introduction
2 Moving Embedded zonal grid method
2.1 Governing Equations
2.2 Moving Grid Finite Volume Method
2.3 Moving Embedded Zonal Grid Method
3 Numerical results
3.1 Validation 1: Geometric Conservation Laws
3.2 Validation 2: Computation of Incompressible Flow-Field Driven by Moving Cube in the Stationary Fluid
3.3 Computation of Incompressible Flow-Field Driven by Moving Twin Cubes in the Stationary Fluid
4 Parallel implementation
5 Concluding remarks
Part IV Boundary methods
Flow Computations Using Embedded Boundary Conditions on Block Structured Cartesian Grid
1 Introduction
2 Building-Cube Method
3 Governing Equations
4 Numerical Method
4.1 Boundary Conditions
4.2 Solution Algorithm
4.3 Parallelization
5 Numerical Results
6 Conclusion
Computation of Two-phase Flow in Flip-chip Packaging Using Level Set Method
1 Numerical Simulation of Capillary Flow Undefill
2 Numerical Results
3 Conclusions
A Parallel Immersed Boundary Method for Blood-like Suspension Flow Simulations
1 The Fluid/elastic-structure interaction equations
1.1 The Context
1.2 The Temporal discretization of the IBM
2 The Parallel flow solver
2.1 The Pressure solver
2.2 The IBM solver
3 The Test-case
4 Conclusion
Part V High Order methods
3D Spectral Parallel Multi-Domain computing for natural convection flows
1 Introduction
2 Basic ideas
3 Physical problem and mathematical equations
4 Numerical schemes and parallel algorithm
5 Code behavior and numerical results
6 Summary
3D time accurate CFD simulations of a centrifugal compressor
1 Introduction
2 Transient effects in turbomachinery
3 Numerical Procedure
4 Results
4.1 Compressor with volute
4.2 Compressor without volute
5 Summary
6 Acknowledgements
Part VI Parallel Algorithms and Solvers
Multicolor SOR Method with Consecutive Memory Access Implementation in a Shared and Distributed Memory Parallel Environment
1 Introduction
2 Multicolor Ordering with Sorting of Memory Layout
2.1 Issues and Concept of Improvement
2.2 Implementation of 2-Colored Ordering SOR with Consecutive Memory Access
2.3 Performance Evaluation
3 Results and Discussion
3.1 Performance on a Shared Memory Vector Computer
3.2 Performance on a Symmetric Multiprocessor Computer
3.3 Performance in a Distributed Memory Parallel Environment
4 Conclusions
Acknowledgement
Proper Orthogonal Decomposition In Decoupling Large Dynamical Systems
1 Introduction
2 Proper Orthogonal Decomposition and Model reduction
2.1 Singular Value Decomposition
2.2 POD in model reduction
2.3 POD for decoupling systems
3 Analysis of POD in solving dynamical system
3.1 Analysis of the reduced model
3.2 Analysis of POD in decoupling systems
4 Parallel implementation and algorithm
5 Numerical tests
Diurnal Kinetic Advection-Diffusion problem
6 Conclusions and future work
Performance Analysis of the Parallel Aitken-Additive Schwarz Waveform Relaxation Method on Distributed Environment
1 Introduction
2 The PAASWR Algorithm
2.1 Definition of the Problem and its Discretization
2.2 General Algorithm
3 Parallel Implementation extended to the Grid
3.1 The Concept of Subdomain Solver and Interface Solver
3.2 The Parallel Non Blocking Windowed Version revisited
4 Performance Results on Distributed Environment
5 Conclusion
Aitken-Schwarz Acceleration not based on the mesh for CFD
1 Introduction
2 Aitken-Schwarz Method for Linear Operators
3 The Aitken acceleration for a sequence of vectors
3.1 The Singular Value Decomposition
4 Numerical results on Darcy equation
5 Conclusions
From extruded-2D to fully-3D geometries for DNS: a Multigrid-based extension of the Poisson solver
1 Introduction
2 Governing equations and numerical method for DNS
3 On the extension of the KSFD algorithm for fully-3D problems
3.1 CG-KSFD approach:
3.2 MG-KSFD approach:
4 Solution of the second MG level
5 Motivation: DNS of a wall-mounted cube
6 Convergence and parallel performance tests
7 Conclusions and future research
Parallel direct Poisson solver for DNS of complex turbulent flows using Unstructured Meshes
1 Introduction
2 Numerical method for DNS
2.1 Spatial discretization of Navier-Stokes equations
2.2 Time integration
3 Poisson solver
3.1 Fourier diagonalization
3.2 Schur Complement Decomposition
4 Numerical Experiments and DNS applications
5 Concluding remarks
A numerical scheme for the computation of phase transition in compressible multiphase flows
1 Thermodynamic model
2 Solution of the Riemann problem
2.1 The initial point is a gas
2.2 The initial point is a mixture
2.3 The initial point is a liquid
3 Numerical scheme
3.1 Numerical scheme
3.2 Second order extension
4 Numerical results
4.1 Validation test 1: split rarefaction wave
4.2 Validation test 2: Liu solution of a split shock
4.3 2 dimensional test
5 Conclusion
Part VII Lattice Boltzman and SPH Methods
Lattice Boltzmann Simulations of Slip Flow of Non-Newtonian Fluids in Microchannels
1 Introduction
2 Brief review of basic theory of laticce Boltzmann method
3 Non-newtonian fluid models
4 Results
4.1 Newtonian and Non-Newtonian Flow in a Microcavity
4.2 No-Slip and Slip Flow of Non-Newtonian Fluids in a Microchannel
5 Parallelization
6 Conclusions
Multiple Relaxation Time Lattice Boltzmann simulation of binary droplet collisions
1 Introduction
2 The Lattice Boltzmann Method for single and multiphase flows
3 Results
4 Discussion and Conclusions
High-Performance Computing and Smoothed Particle Hydrodynamics
1 Introduction
2 Methodology
3 Equations in SPH framework
4 Parallelisation
4.1 Step 1. Generation of the new particle list
4.2 Step 2. Search for particle links
4.3 Step 3. Link particle re-indexing
4.4 Step 4. Resolution of the equations
5 Results
5.1 Description of the Machines
5.2 Accuracy
5.3 Performance of the Code
6 Concluding remarks
Part VIII software Framework and Component Architecture
An integrated object-oriented approach for parallel CFD
1 Introduction
2 Java in Computational Mechanics
3 Schwarz multiplicative multithreaded and distributed applications
4 Numerical application
5 Conclusion
Fast Multipole Method for particle interactions: an open source parallel library component
1 INTRODUCTION
2 CHARACTERIZATION OF THE MULTIPOLE APPROXIMATION
2.1 Overview of the algorithm
3 PARALLELIZATION STRATEGY
4 RESULTS
5 CONCLUSIONS AND FUTURE WORK
Hybrid MPI-OpenMP performance in massively parallel computational fluid dynamics
1 Introduction
2 Incompressible flow equations
3 Compressible flow equations
4 Parallelization strategy
5 Performance
Hierarchical adaptive multi-mesh partitioning algorithm on heterogeneous systems
1 Introduction
2 The iterative scheduling algorithm
3 Partitioning Problem
4 Hierarchical partitioning strategy
5 Experimentation of a finite element code on the grid
5.1 Pour of a jerrican
6 Conclusion
Part IX Parallel Performance
Towards Petascale Computing with Parallel CFD codes
1 Introduction
2 Parallel CFD codes
2.1 SBLI
2.2 SENGA2
2.3 STREAMS-LES
2.4 CODE SATURNE
3 High-End Computing Platforms
3.1 HPCx
3.2 HECToR
3.3 BlueGene
4 Performance Results
4.1 SBLI
4.2 STREAMS-LES
4.3 Code Saturne
4.4 SENGA2
5 Summary
Scalability Considerations of a Parallel Flow Solver on Large Computing Systems
1 Introduction
2 Flow Solver
2.1 Features of the Flow Solver
2.2 Overset Grid Method
2.3 Comparison of Two Flow Solvers
3 Numerical Simulations
3.1 Parallel CFD Application
3.2 Timing Analysis by Vampir Tool
3.3 Performance of the Overset Version
4 Conclusions
Large Scaled Computation of Incompressible Flows on Cartesian Mesh Using a Vector-Parallel Supercomputer
1 Introduction
2 Computational methods
2.1 Building-Cube Method
2.2 Vectorization and parallelization
3 Results and discussion
3.1 Computational mesh and result
3.2 Parallel efficiency
4 Conclusions
Dynamic Load Balancing on Networked Multi-core Computers
1 Introduction
2 Dynamic Load Balancing
3 DLB Implementation
4 Experiments
5 Conclusion
On efficiency of supercomputers in CFD simulations
1 Introduction
2 Supercomputers and codes under consideration
3 Efficiency of multi-core nodes
4 The two-level preprocessing model
5 Hybrid parallelization based on MPI and OpenMP
6 Performance and illustrative results
Part X Environment and biofluids applications
Numerical Study of Pulsatile Flow Through Models of Vascular Stenoses with Physiological Waveform of the Heart
1 Introduction
2 CFD Fluent solver
3 Results
3.1 Pulsatile Flow in Axisymmetric Stenosed Phantoms
3.2 Pulsatile Flow in Asymmetric Stenosed Phantoms
4 Parallelisation
5 Conclusions
Fluid Flow - Agent Based Hybrid Model for the Simulation of Virtual Prairies
1 Problem Background
2 Hybrid Model
3 Computation Needs for Virtual Prairie
4 Virtual Prairie with BOINC
5 Conclusion
HPC for hydraulics and industrial environmental flow simulations
1 HPC through the TELEMAC system
1.1 Historic of HPC in the TELEMAC system
1.2 TELEMAC, a SPMD hydroinformatics system
1.3 Structure of parallelism in the TELEMAC system
1.4 Point to point communication and data exchange
2 HPC through soil pollution with Estel-3D
2.1 Estel-3D code
2.2 Application of Estel-3D in the vadose zone
3 HPC for real environmental applications with Telemac-3D
3.1 Telemac-3D code
3.2 Application to Berre lagoon
4 HPC for complex environmental flows with Spartacus-3D
4.1 The SPH method
4.2 Spartacus-3D code
4.3 Application to an industrial hydraulic structure
5 Conclusion
Multi-parametric intensive stochastic simulations for hydrogeology on a computational grid
1 Introduction
2 Existing work
2.1 Parallel Monte-Carlo
2.2 Distributed multiple simulations and grid applications
2.3 Middleware on grids
3 Our work
3.1 Generic tools in H2OLab
3.2 Deployment on grid architectures
3.3 Experiments on clusters and conclusion
Parallel computation of pollutant dispersion in industrial sites
1 Description of the numerical method
1.1 Implicit Navier Stokes Solver
1.2 Iterative methods for sparse linear systems
Coarsening schemes
Interpolation
2 Description of the tests
2.1 Laplace equation
2.2 Flow around a complex industrial site
3 Results
3.1 Laplace equation
3.2 Flow around a complex industrial site
4 Conclusion
Part XI General fluid
3D Numerical Simulation Of Gas Flow Around Reentry Vehicles
1 Introduction
2 Mathematical Model
3 Numerical Implementation
4 Parallelization
5 Test case
6 Conclusions
7 Acknowledgments
Effective Parallel Computation of Incompressible Turbulent Flows on Non-uniform Grid
1 Introduction
2 Computational Technique
2.1 Variable Order Method of Lines
2.2 Pressure Equation Solver
3 Numerical Results
3.1 3D Test Problem
3.2 DNS of 3D Turbulent Channel Flow
4 Concluding Remarks
Secondary flow structure of turbulent Couette-Poiseuille and Couette flows inside a square duct
1 Introduction
2 Governing Equations and Modeling
3 Numerical and parallel Algorithms
4 Results
4.1 Mean secondary flow structure
4.2 Anisotropy invariant map
5 Conclusion
6 Acknowledgments
Editorial Policy
Lecture Notesin Computational Scienceand Engineering
Monographs in Computational Scienceand Engineering
Texts in Computational Scienceand Engineering

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Parallel Computational Fluid Dynamics 2008

Description

More details

Other editions

Additional editions

Content

System requirements