High-Performance Scientific Computing

Name: High-Performance Scientific Computing | First JARA-HPC Symposium, JHPCS 2016, Aachen, Germany, October 4-5, 2016, Revised Selected Papers
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

First JARA-HPC Symposium, JHPCS 2016, Aachen, Germany, October 4-5, 2016, Revised Selected Papers

Edoardo Di Napoli Marc-André Hermanns Hristo Iliev Andreas Lintermann Alexander Peyser(Herausgeber*in)

Springer (Verlag)

Erschienen am 1. März 2017

XV, 258 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-319-53862-4 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Preface
Organization
Invited Talk (Abstract)
Parallel Programming for the 21st Century
Contents
Efficient HPC-Optimized Multi-Physics Coupling Strategies in CFD
Partitioned High Performance Code Coupling Applied to CFD
1 Introduction
2 In Situ Observations
3 Toy Model
4 Conclusion
References
Dynamic Load Balancing for Large-Scale Multiphysics Simulations
1 Introduction
2 Static Load Balancing
3 Dynamic Load Balancing
3.1 Workload Modeling
3.2 Intelligent Remapping
4 Performance Evaluation
4.1 Immersed Boundary Method
4.2 IB Workload Modeling
4.3 Load Balancing Threshold
4.4 Nose Landing Gear
4.5 Full Car Model
5 Discussion
6 Summary and Future Work
References
On the Significance of Exposure Time in Computational Blood Damage Estimation
1 Introduction
2 Methods
3 Results
3.1 Blood Flow Simulations
3.2 Hemolysis Simulations
4 Discussion
5 Conclusion
References
A Partitioned Methodology for Conjugate Heat Transfer on Dynamic Structures
1 Introduction
2 Mathematical Modelling
2.1 Governing Equations
2.2 Conjugate Heat Transfer
3 Numerical Methodology
3.1 Finite Element Method
3.2 Coupling Approach
4 Numerical Results
4.1 Heated Cylinder
4.2 Flat-Plate
4.3 Parallel Efficiency
5 Conclusions and Future Work
References
Farfield Noise Prediction Using Large-Scale Lattice-Boltzmann Simulations
Abstract
1 Introduction
2 Numerical Method
3 Simulations
3.1 Isolated Nose Landing Gear
3.2 Full-Scale Aircraft in Landing Configuration
4 Conclusion
References
FEniCS-HPC: Coupled Multiphysics in Computational Fluid Dynamics
1 Introduction
2 The FEniCS-HPC Framework
3 Mathematical Model and Numerical Methods
3.1 Conservation Equations
3.2 Fluid-Structure Interaction
3.3 Finite Element Approximation for the Pure Fluid Case
3.4 Turbulent Flow
3.5 Contact Model
3.6 Mesh Smoothing Algorithms
4 Multiphysics in Computational Fluid Dynamics
4.1 Heart Biomechanics
4.2 Human Phonation
4.3 Aeroacoustics
References
The Direct-Hybrid Method for Computational Aeroacoustics on HPC Systems
1 Introduction
2 The Direct-Hybrid Method
3 Governing Equations
4 Numerical Methods
5 Results
6 Conclusions
References
A Novel Approach for Efficient Storage and Retrieval of Tabulated Chemistry in Reactive Flow Simulations
1 Introduction
2 Flatkernel Approach
2.1 Flatkernel Generation
2.2 Flatkernel Access
3 Reactive Hydrogen Jet in a Turbulent Vitiated Cross Flow
3.1 Case Description
3.2 Results and Discussion
4 Summary
References
Multi-scale Coupling for Predictive Injector Simulations
1 Introduction
2 Numerical Framework
3 Application Setup and Coupling
3.1 ``Spray A'' Case
3.2 Coupling Interface Between Nozzle Internal Flow and Primary Breakup
3.3 Coupling Interface Between Primary Breakup and Secondary Breakup
4 Results and Discussion
5 Conclusions
References
Domain-Specific Applications and High-Performance Computing
Ab Initio Description of Optoelectronic Properties at Defective Interfaces in Solar Cells
1 Introduction
2 Method
2.1 Atomic Structure Calculations
2.2 Electronic Structure Calculations
2.3 Optical Calculations
3 Results
3.1 A-Si:H
3.2 C-Si/a-Si:H Interface
4 Computational Costs
5 Conclusions
References
Scale Bridging Simulations of Large Elastic Deformations and Bainitic Transformations
1 Introduction
2 Nonlinear Elasticity in Phase Field Crystal Models and Comparison to ab initio Simulations
2.1 Modeling Approach
2.2 One-Dimensional Nonlinear Elasticity
2.3 Three-Dimensional Nonlinear Elasticity for bcc Materials
2.4 Comparison with ab initio Simulations and Continuum Modeling
3 Phase Field Model of Bainite Formation
3.1 Multi Phase Field Modeling
3.2 Results
4 Near-Tip Segregation at a Mode I Crack
5 GPU Implementation
6 Conclusions
References
Ab Initio Modelling of Electrode Material Properties
1 Introduction
2 Methods
3 Results
4 Summary and conclusions
References
Overlapping of Communication and Computation in nb3dfft for 3D Fast Fourier Transformations
1 Introduction
2 Pseudo-spectral Algorithm
3 Overlapping Communication and Computation
4 Results
5 Conclusion
References
Towards Simulating Data-Driven Brain Models at the Point Neuron Level on Petascale Computers
1 Introduction
2 Contribution
3 Methods
3.1 PointBrainH5
3.2 Implementation
3.3 RTC Algorithm
4 Performance Results
5 Functionality
6 Summary
References
Parallel Adaptive Integration in High-Performance Functional Renormalization Group Computations
1 Introduction
2 The fRG Method and the Truncated Unity Approach
3 Adaptive Integration A-La Clenshaw-Curtis
4 Parallel Implementation
5 Results and Conclusions
References
Performance Portability
Performance Optimization of Parallel Applications in Diverse On-Demand Development Teams
1 Introduction
2 Mission Statement
2.1 Related Activities
3 Case Study: psOpen
4 Case Study: MP2C
5 Conclusion
References
Hybrid CPU-GPU Generation of the Hamiltonian and Overlap Matrices in FLAPW Methods
1 Introduction
1.1 DFT, FLAPW and the H and S Matrices
2 Algorithm
3 Software Re-engineering and Performance Portability
4 Experimental Results
4.1 Fine-Tuning for Performance and Scalability
5 Conclusions and Future Work
References
Visualizing Performance Data with Respect to the Simulated Geometry
1 Introduction
2 Related Work
3 Nomenclature: Performance Profiles, Severity Views
4 Detecting Large-Impact, Large-Variation Views
5 Interactive Visualizations
5.1 Performance Metrics and Call-Path Tree Widgets
5.2 Visualizing Performance Data in Its Spatial Context
6 Results
7 Conclusion and Future Work
References
Provenance Tracking
Framework for Sharing of Highly Resolved Turbulence Simulation Data
1 Introduction
2 State-of-the-Art and Motivation
3 Highly Resolved Turbulence Simulation
4 Data Sharing Framework
5 Conclusion
References
UniProv: A Flexible Provenance Tracking System for UNICORE
1 Introduction
2 Workflows and Provenance
2.1 Scientific Workflow Systems
2.2 Related Work
2.3 Neuroimaging Workflows with UNICORE
3 UniProv Provenance Management
3.1 Implementation
3.2 Flexible Extensibility
3.3 Provenance Repository
4 Conclusion
5 Future Work
References
A Collaborative Simulation-Analysis Workflow for Computational Neuroscience Using HPC
1 Introduction
2 Workflow
3 Discussion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken