Search

Your search keyword '"Christian Obrecht"' showing total 39 results
Start Over Author "Christian Obrecht"
39 results on '"Christian Obrecht"'

1. Artificial Neural Network Simulation of Energetic Performance for Sorption Thermal Energy Storage Reactors

Author

Carla Delmarre, Marie-Anne Resmond, Frédéric Kuznik, Christian Obrecht, Bao Chen, and Kévyn Johannes

Subjects
sorption thermal energy storage, zeolite, artificial neural network, recurrent neural network, Technology
Abstract

Sorption thermal heat storage is a promising solution to improve the development of renewable energies and to promote a rational use of energy both for industry and households. These systems store thermal energy through physico-chemical sorption/desorption reactions that are also termed hydration/dehydration. Their introduction to the market requires to assess their energy performances, usually analysed by numerical simulation of the overall system. To address this, physical models are commonly developed and used. However, simulation based on such models are time-consuming which does not allow their use for yearly simulations. Artificial neural network (ANN)-based models, which are known for their computational efficiency, may overcome this issue. Therefore, the main objective of this study is to investigate the use of an ANN model to simulate a sorption heat storage system, instead of using a physical model. The neural network is trained using experimental results in order to evaluate this approach on actual systems. By using a recurrent neural network (RNN) and the Deep Learning Toolbox in MATLAB, a good accuracy is reached, and the predicted results are close to the experimental results. The root mean squared error for the prediction of the temperature difference during the thermal energy storage process is less than 3K for both hydration and dehydration, the maximal temperature difference being, respectively, about 90K and 40K.

Published
2021
Full Text
View/download PDF

7. Query-adaptive training data recommendation for cross-building predictive modeling

Author

Mouna Labiadh, Christian Obrecht, Catarina Ferreira da Silva, Parisa Ghodous, and Khalid Benabdeslem

Subjects
Human-Computer Interaction, Artificial Intelligence, Hardware and Architecture, Data-driven modeling, Similarity learning, Training data recommendation, Knowledge transfer, Ciências Naturais::Ciências da Computação e da Informação [Domínio/Área Científica], Building energy, Domain generalization, Software, Information Systems
Abstract

Predictive modeling in buildings is a key task for the optimal management of building energy. Relevant building operational data are a prerequisite for such task, notably when deep learning is used. However, building operational data are not always available, such is the case in newly built, newly renovated, or even not yet built buildings. To address this problem, we propose a deep similarity learning approach to recommend relevant training data to a target building solely by using a minimal contextual description on it. Contextual descriptions are modeled as user queries. We further propose to ensemble most used machine learning algorithms in the context of predictive modeling. This contributes to the genericity of the proposed methodology. Experimental evaluations show that our methodology offers a generic methodology for cross-building predictive modeling and achieves good generalization performance. info:eu-repo/semantics/acceptedVersion

Published
2022
Full Text
View/download PDF

19. A microservice-based framework for exploring data selection in cross-building knowledge transfer

Author

Mouna Labiadh, Christian Obrecht, Parisa Ghodous, Catarina Ferreira da Silva, Service Oriented Computing (SOC), Laboratoire d'InfoRmatique en Image et Systèmes d'information (LIRIS), Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-École Centrale de Lyon (ECL), Université de Lyon-Université Lumière - Lyon 2 (UL2)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Université Lumière - Lyon 2 (UL2), Centre d'Energétique et de Thermique de Lyon (CETHIL), Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon, and Instituto Universitário de Lisboa (ISCTE-IUL)

Subjects
FOS: Computer and information sciences, Computer Science - Machine Learning, Generalization, Computer science, [INFO.INFO-DS]Computer Science [cs]/Data Structures and Algorithms [cs.DS], Knowledge transfer, Context (language use), data-driven-modeling, 02 engineering and technology, Machine learning, computer.software_genre, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Machine Learning (cs.LG), Computer Science - Information Retrieval, Management Information Systems, Domain (software engineering), Data selection, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Computer Science - Databases, Data-driven modeling, 020204 information systems, 0202 electrical engineering, electronic engineering, information engineering, Selection (linguistics), energy consumption modeling, Engenharia e Tecnologia::Outras Engenharias e Tecnologias [Domínio/Área Científica], Domain generalization, Energy consumption modeling, [INFO.INFO-DB]Computer Science [cs]/Databases [cs.DB], Data collection, business.industry, Deep learning, Ciências Naturais::Ciências da Computação e da Informação [Domínio/Área Científica], Databases (cs.DB), 020207 software engineering, Energy consumption, [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation, Hardware and Architecture, [INFO.INFO-IR]Computer Science [cs]/Information Retrieval [cs.IR], Artificial intelligence, business, computer, Information Retrieval (cs.IR), Software, Information Systems
Abstract

Supervised deep learning has achieved remarkable success in various applications. Successful machine learning application however depends on the availability of sufficiently large amount of data. In the absence of data from the target domain, representative data collection from multiple sources is often needed. However, a model trained on existing multi-source data might generalize poorly on the unseen target domain. This problem is referred to as domain shift. In this paper, we explore the suitability of multi-source training data selection to tackle the domain shift challenge in the context of domain generalization. We also propose a microservice-oriented methodology for supporting this solution. We perform our experimental study on the use case of building energy consumption prediction. Experimental results suggest that minimal building description is capable of improving cross-building generalization performances when used to select energy consumption data., Service Oriented Computing and Applications, Springer, 2020

Published
2020
Full Text
View/download PDF

20. Artificial Neural Network Simulation of Energetic Performance for Sorption Thermal Energy Storage Reactors

Author

Bao Chen, Marie-Anne Resmond, Christian Obrecht, Carla Delmarre, Frédéric Kuznik, and Kévyn Johannes

Subjects
sorption thermal energy storage, zeolite, artificial neural network, recurrent neural network, Technology, Control and Optimization, Computer science, 020209 energy, Energy Engineering and Power Technology, 02 engineering and technology, Thermal energy storage, 0202 electrical engineering, electronic engineering, information engineering, Electrical and Electronic Engineering, Process engineering, Engineering (miscellaneous), Computer simulation, Artificial neural network, Renewable Energy, Sustainability and the Environment, business.industry, Deep learning, Sorption, 021001 nanoscience & nanotechnology, Renewable energy, Recurrent neural network, Computer Science::Programming Languages, Artificial intelligence, 0210 nano-technology, business, Thermal energy, Energy (miscellaneous)
Abstract

Sorption thermal heat storage is a promising solution to improve the development of renewable energies and to promote a rational use of energy both for industry and households. These systems store thermal energy through physico-chemical sorption/desorption reactions that are also termed hydration/dehydration. Their introduction to the market requires to assess their energy performances, usually analysed by numerical simulation of the overall system. To address this, physical models are commonly developed and used. However, simulation based on such models are time-consuming which does not allow their use for yearly simulations. Artificial neural network (ANN)-based models, which are known for their computational efficiency, may overcome this issue. Therefore, the main objective of this study is to investigate the use of an ANN model to simulate a sorption heat storage system, instead of using a physical model. The neural network is trained using experimental results in order to evaluate this approach on actual systems. By using a recurrent neural network (RNN) and the Deep Learning Toolbox in MATLAB, a good accuracy is reached, and the predicted results are close to the experimental results. The root mean squared error for the prediction of the temperature difference during the thermal energy storage process is less than 3K for both hydration and dehydration, the maximal temperature difference being, respectively, about 90K and 40K.

Published
2021
Full Text
View/download PDF

21. On the suitability of Data Selection for Cross-building Knowledge Transfer

Author

Christian Obrecht, Mouna Labiadh, Catarina Ferreira da Silva, Parisa Ghodous, Service Oriented Computing (SOC), Laboratoire d'InfoRmatique en Image et Systèmes d'information (LIRIS), Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-École Centrale de Lyon (ECL), Université de Lyon-Université Lumière - Lyon 2 (UL2)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Université Lumière - Lyon 2 (UL2), Université Claude Bernard Lyon 1 - Faculté des sciences et technologies (UCBL FST), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon, Centre d'Energétique et de Thermique de Lyon (CETHIL), Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université Claude Bernard Lyon 1 (UCBL), and Université de Lyon

Subjects
Generalization, Computer science, [SHS.INFO]Humanities and Social Sciences/Library and information sciences, Context (language use), 02 engineering and technology, 010501 environmental sciences, Machine learning, computer.software_genre, 01 natural sciences, Domain (software engineering), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Data selection, 0202 electrical engineering, electronic engineering, information engineering, energy consumption modeling, [INFO]Computer Science [cs], domain generalization, ComputingMilieux_MISCELLANEOUS, 0105 earth and related environmental sciences, Training set, Data collection, business.industry, Deep learning, knowledge transfer, data-driven modeling, 020201 artificial intelligence & image processing, Artificial intelligence, business, Knowledge transfer, computer
Abstract

Supervised deep learning has achieved remarkable success in various applications. Such advances were mainly attributed to the rise of computational powers and the amounts of training data made available. Therefore, accurate large-scale data collection services are often needed. Once representative data is retrieved, it becomes possible to train the supervised machine learning predictor. However, a model trained on existing data, that generally comes from multiple datasets, might generalize poorly on the unseen target data. This problem is referred to as domain shift. In this paper, we explore the suitability of data selection to tackle the domain shift challenge in the context of domain generalization. We perform our experimental study on the use case of building energy consumption prediction. Experimental results suggest that minimal building description is capable of improving cross-building generalization performances when used to select data.

Published
2019

22. Sorption materials characterization and lab-scale reaction

Author

Margalida, Fullana-Puig, Valeria, Palomba, Atef, Salem, Rosa, Mondragón, Fernandez, Angel G., Aran, Solé, Calabrese, Luigi, Christian, Obrecht, Leonor, Hernández, Andrea, Frazzica, and Cabeza, Luisa F.

Subjects
salt hydrate, metal corrosion, thermal conductivity, TES, sorption material, salt hydrate, metal corrosion, thermal conductivity, TES, sorption material
Published
2019

23. A review on recent developments in physisorption thermal energy storage for building applications

Author

Christian Obrecht, Damien David, Kévyn Johannes, Frédéric Kuznik, Centre d'Energétique et de Thermique de Lyon (CETHIL), Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université Claude Bernard Lyon 1 (UCBL), and Université de Lyon

Subjects
Renewable Energy, Sustainability and the Environment, business.industry, Process (engineering), 020209 energy, Scale (chemistry), 02 engineering and technology, Technology readiness level, Thermal energy storage, [SPI]Engineering Sciences [physics], Physisorption, 0202 electrical engineering, electronic engineering, information engineering, Energy density, Environmental science, Process engineering, business, Energy (signal processing), Thermal energy, ComputingMilieux_MISCELLANEOUS
Abstract

On one hand, physical adsorption, also named physisorption, is a process that can be used to storage thermal energy with an energy density higher than sensible or latent storages. On the other hand, in Europe, 26% of the final energy consumption is related to the energy systems of households [1] and 80% of this energy is needed for heating purposes [2] . The consequence is the development of thermal energy storage systems, based on physisoprtion, for building application. The objective of this paper is first to present the basics concerning physisorption heat storage. Then, experimental developments from the literature are reviewed, based on three scales: the material scale, the reactor scale and the system scale. From the review, development of commercial systems faces with scientific and technological issues that must be addressed to reach a higher technology readiness level with an acceptable system cost.

Published
2018
Full Text
View/download PDF

24. Lattice Boltzmann model approximated with finite difference expressions

Author

Christian Obrecht, François Dubois, Pierre Lallemand, Mohamed Mahdi Tekitek, Laboratoire de Mathématiques d'Orsay (LM-Orsay), Université Paris-Sud - Paris 11 (UP11)-Centre National de la Recherche Scientifique (CNRS), Laboratoire de Mécanique des Structures et des Systèmes Couplés (LMSSC), Conservatoire National des Arts et Métiers [CNAM] (CNAM), Beijing Computational Science Research Center [Beijing] (CSRC), Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA), Université de Tunis El Manar (UTM), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), and Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)

Subjects
quartic parameters, General Computer Science, Lattice boltzmann model, Lattice Boltzmann methods, FOS: Physical sciences, 010103 numerical & computational mathematics, 01 natural sciences, Physics::Fluid Dynamics, FOS: Mathematics, Mathematics - Numerical Analysis, 0101 mathematics, Mathematics, artificial compressibility method, [SPI.FLUID]Engineering Sciences [physics]/Reactive fluid environment, Mathematical analysis, Fluid Dynamics (physics.flu-dyn), General Engineering, Finite difference, Numerical Analysis (math.NA), Physics - Fluid Dynamics, Computational Physics (physics.comp-ph), Hagen–Poiseuille equation, First order, 010101 applied mathematics, Nonlinear system, Physics - Computational Physics, [MATH.MATH-NA]Mathematics [math]/Numerical Analysis [math.NA]
Abstract

International audience; We show that the asymptotic properties of the link-wise artificial com-pressibility method are not compatible with a correct approximation of fluid properties. We propose to adapt the previous method through a framework suggested by the Tay-lor expansion method and to replace first order terms in the expansion by appropriate three or five points finite differences and to add non linear terms. The " FD-LBM " scheme obtained by this method is tested in two dimensions for shear wave, Stokes modes and Poiseuille flow. The results are compared with the usual lattice Boltzmann method in the framework of multiple relaxation times.

Published
2018
Full Text
View/download PDF

25. Design and characterisation of a high powered energy dense zeolite thermal energy storage system for buildings

Author

Francois Durier, Christian Obrecht, Jean-Luc Hubert, Frédéric Kuznik, Kévyn Johannes, Centre d'Energétique et de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), EDF (EDF), CETIAT, and Centre Technique des Industries Aérauliques et Thermiques

Subjects
Materials science, Hygrometer, business.industry, Mechanical Engineering, Nuclear engineering, Airflow, Mechanical engineering, Building and Construction, Management, Monitoring, Policy and Law, Thermal energy storage, Volumetric flow rate, General Energy, Computer data storage, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], Relative humidity, business, ComputingMilieux_MISCELLANEOUS, Water vapor, Thermal energy
Abstract

This paper presents the design and the characterisation of a high powered energy dense zeolite thermal heat storage system using water vapour sorbate. The specification requirements of the system are to supply a heating power of 2 kW during 2 h in order to shave the electricity peak loads in a house. The open reactor has been designed, built and instrumented with temperature sensors, chilled mirror hygrometer and airflow meter. Several tests have been carried out both during hydration – heat release – and dehydration – heat storage. Tests have also been carried out for different flow rates, relative humidity and temperatures of hydration. The results show that the reactor can supply a constant power of 2.25 kW during more than two hours, namely 27.5 W kg−1 of material.

Published
2015
Full Text
View/download PDF

26. Hybrid thermal link-wise artificial compressibility method

Author

Frédéric Kuznik, Christian Obrecht, Centre d'Energétique et de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), and Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)

Subjects
Convection, Physics, Natural convection, Finite difference, General Physics and Astronomy, Thermodynamics, Mechanics, Nusselt number, Isothermal process, Physics::Fluid Dynamics, Flow (mathematics), Thermal, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], Compressibility, ComputingMilieux_MISCELLANEOUS
Abstract

Thermal flow prediction is a subject of interest from a scientific and engineering points of view. Our motivation is to develop an accurate, easy to implement and highly scalable method for convective flows simulation. To this end, we present an extension to the link-wise artificial compressibility method (LW-ACM) for thermal simulation of weakly compressible flows. The novel hybrid formulation uses second-order finite difference operators of the energy equation based on the same stencils as the LW-ACM. For validation purposes, the differentially heated cubic cavity was simulated. The simulations remained stable for Rayleigh numbers up to Ra = 10 8 . The Nusselt numbers at isothermal walls and dynamics quantities are in good agreement with reference values from the literature. Our results show that the hybrid thermal LW-ACM is an effective and easy-to-use solution to solve convective flows.

Published
2015
Full Text
View/download PDF

27. High-performance implementations and large-scale validation of the link-wise artificial compressibility method

Author

Pietro Asinari, Jean-Jacques Roux, Frédéric Kuznik, Christian Obrecht, Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), Multi-scale Modeling Lab (SMaLL), Politecnico di Torino [Torino] (Polito), CETHIL, Laboratoire, Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), and Politecnico di Torino = Polytechnic of Turin (Polito)

Subjects
Physics and Astronomy (miscellaneous), Computer science, Reference data (financial markets), Lattice Boltzmann methods, CUDA, 010103 numerical & computational mathematics, Computational fluid dynamics, 01 natural sciences, Lid-driven cubic cavity, 010305 fluids & plasmas, Computational science, symbols.namesake, 0103 physical sciences, 0101 mathematics, Link-wise artificial compressibility method, High-performance computing, [SPI.MECA.THER] Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], Physics::Computational Physics, Numerical Analysis, [PHYS.MECA.THER] Physics [physics]/Mechanics [physics]/Thermics [physics.class-ph], business.industry, Applied Mathematics, Reynolds number, Solver, Nonlinear Sciences::Cellular Automata and Lattice Gases, Supercomputer, Computer Science Applications, Computer Science::Performance, Computational Mathematics, Modeling and Simulation, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], [PHYS.MECA.THER]Physics [physics]/Mechanics [physics]/Thermics [physics.class-ph], symbols, Compressibility, business
Abstract

International audience; The link-wise artificial compressibility method (LW-ACM) is a recent formulation of the artificial compressibility method for solving the incompressible Navier-Stokes equations. Two implementations of the LW-ACM in three dimensions on CUDA enabled GPUs are described. The first one is a modified version of a state-of-the-art CUDA implementation of the lattice Boltzmann method (LBM), showing that an existing GPU LBM solver might easily be adapted to LW-ACM. The second one follows a novel approach, which leads to a performance increase of up to 1.8× compared to the LBM implementation considered here, while reducing the memory requirements by a factor of 5.25. Large-scale simulations of the lid-driven cubic cavity at Reynolds number Re=2000Re=2000 were performed for both LW-ACM and LBM. Comparison of the simulation results against spectral elements reference data shows that LW-ACM performs almost as well as multiple-relaxation-time LBM in terms of accuracy.

Published
2014
Full Text
View/download PDF

28. Multi-GPU implementation of a hybrid thermal lattice Boltzmann solver using the TheLMA framework

Author

Frédéric Kuznik, Christian Obrecht, Bernard Tourancheau, Jean-Jacques Roux, Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), and Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)

Subjects
Coupling, General Computer Science, Computer science, General Engineering, Lattice Boltzmann methods, CUDA, GPU computing, Solver, 01 natural sciences, 010305 fluids & plasmas, Computational science, 010101 applied mathematics, Thermal lattice Boltzmann method, Component (UML), 0103 physical sciences, Thermal, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], [PHYS.MECA.THER]Physics [physics]/Mechanics [physics]/Thermics [physics.class-ph], Fluid dynamics, 0101 mathematics, General-purpose computing on graphics processing units
Abstract

International audience; In this contribution, a single-node multi-GPU thermal lattice Boltzmann solver is presented. We implement a simplified version of the hybrid model developed by Lallemand and Luo in 2003, which combines multiple-relaxation-time lattice Boltzmann for the fluid flow with a finite-difference method for temperature. The program is based on the TheLMA framework which was developed for that purpose. The chosen implementation and optimisation strategies are described, both for inter-GPU communication and for coupling with the thermal component of the model. Validation and performance results are provided as well.

Published
2013
Full Text
View/download PDF

29. Efficient GPU implementation of the linearly interpolated bounce-back boundary condition

Author

Frédéric Kuznik, Christian Obrecht, Bernard Tourancheau, Jean-Jacques Roux, Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), and Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)

Subjects
Lattice Boltzmann method, Lattice Boltzmann methods, CUDA, GPU programming, TheLMA project, Parallel computing, 01 natural sciences, 010305 fluids & plasmas, Computational science, Physics::Fluid Dynamics, symbols.namesake, Modelling and Simulation, 0103 physical sciences, Boundary value problem, 010306 general physics, ComputingMethodologies_COMPUTERGRAPHICS, Mathematics, Interpolated bounce-back boundary condition, Reynolds number, Solver, Computational Mathematics, Computational Theory and Mathematics, Flow (mathematics), Modeling and Simulation, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], [PHYS.MECA.THER]Physics [physics]/Mechanics [physics]/Thermics [physics.class-ph], symbols, Node (circuits), General-purpose computing on graphics processing units
Abstract

International audience; Interpolated bounce-back boundary conditions for the lattice Boltzmann method (LBM) make the accurate representation of complex geometries possible. In the present work, we describe an implementation of a linearly interpolated bounce-back (LIBB) boundary condition for graphics processing units (GPUs). To validate our code, we simulated the flow past a sphere in a square channel. At low Reynolds numbers, results are in good agreement with experimental data. Moreover, we give an estimate of the critical Reynolds number for transition from steady to periodic flow. Performance recorded on a single node server with eight GPU based computing devices ranged up to 2.63×1092.63×109 fluid node updates per second. Comparison with a simple bounce-back version of the solver shows that the impact of LIBB on performance is fairly low.

Published
2013
Full Text
View/download PDF

31. Chemisorption heat storage in buildings: state-of-the-art and outlook

Author

Frédéric Kuznik, Kévyn Johannes, Christian Obrecht, Centre d'Energétique et de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), and Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)

Subjects
Work (thermodynamics), Waste management, business.industry, Mechanical Engineering, Renewable heat, Building and Construction, Thermal energy storage, Water production, Chemisorption, Energy density, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], Environmental science, State (computer science), Electrical and Electronic Engineering, Process engineering, business, ComputingMilieux_MISCELLANEOUS, Civil and Structural Engineering, Copper in heat exchangers
Abstract

Chemisorption is a promising solution for long-term heat storage with a high energy density. Considering application to buildings, heat storage can be used for space heating and domestic hot water production. In the present work, the specification requirements of a heat storage system for buildings are given. A review on chemical heat storage materials is carried out and reactor development is discussed. Some future research topics are presented in the conclusions.

Published
2015
Full Text
View/download PDF

32. Towards aeraulic simulations at urban scale using the lattice Boltzmann method

Author

Lucie Merlier, Jean-Jacques Roux, Bernard Tourancheau, Christian Obrecht, Frédéric Kuznik, Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), Drakkar, Laboratoire d'Informatique de Grenoble (LIG), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA), and Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP)-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP)-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)

Subjects
Scale (ratio), Computer science, business.industry, Lattice Boltzmann methods, Solver, Computational fluid dynamics, Supercomputer, Computational science, Computational physics, Physics::Fluid Dynamics, Flow (mathematics), [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], [PHYS.MECA.THER]Physics [physics]/Mechanics [physics]/Thermics [physics.class-ph], Environmental Chemistry, Cube, business, Water Science and Technology, Large eddy simulation
Abstract

International audience; The lattice Boltzmann method (LBM) is an innovative approach in computational fluid dynamics (CFD). Due to the underlying lattice structure, the LBM is inherently parallel and therefore well suited for high performance computing. Its application to outdoor aeraulic studies is promising, e.g. applied on complex urban configurations, as an alternative approach to the commonplace Reynolds-averaged Navier-Stokes and large eddy simulation methods based on the Navier-Stokes equations. Emerging many-core devices, such as graphic processing units (GPUs), nowadays make possible to run very large scale simulations on rather inexpensive hardware. In this paper, we present simulation results obtained using our multi-GPU LBM solver. For validation purpose, we study the flow around a wall-mounted cube and show agreement with previously published experimental results. Furthermore, we discuss larger scale flow simulations involving nine cubes which demonstrate the practicability of CFD simulations in building external aeraulics.

Published
2015
Full Text
View/download PDF

33. Towards high-performance thermal flow solvers based on the link-wise artificial compressibility method

Author

Gilles Rusaouen, Christian Obrecht, Frédéric Kuznik, Jean-Jacques Roux, Centre d'Energétique et de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), and Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS)

Subjects
Mathematical optimization, Artificial compressibility, Flow (mathematics), Computer science, Thermal, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], General-purpose computing on graphics processing units, Link (knot theory), ComputingMilieux_MISCELLANEOUS, Computational science
Abstract

International audience

Published
2014
Full Text
View/download PDF

34. Scalable Lattice Boltzmann Solvers for CUDA GPU Clusters

Author

Bernard Tourancheau, Frédéric Kuznik, Christian Obrecht, Jean-Jacques Roux, Centre de Thermique de Lyon (CETHIL), Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA), Drakkar, Laboratoire d'Informatique de Grenoble (LIG), Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP)-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP)-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), and Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)

Subjects
Computer Networks and Communications, Computer science, Computation, Lattice Boltzmann method, Lattice Boltzmann methods, CUDA, 02 engineering and technology, Parallel computing, Computational fluid dynamics, Theoretical Computer Science, Computational science, Domain (software engineering), [INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI], Artificial Intelligence, 0202 electrical engineering, electronic engineering, information engineering, Throughput (business), ComputingMethodologies_COMPUTERGRAPHICS, Physics::Computational Physics, 020203 distributed computing, business.industry, GPU cluster, Solver, Computer Graphics and Computer-Aided Design, Computer Science::Performance, GPU clusters, Computer Science::Graphics, Hardware and Architecture, Scalability, Computer Science::Mathematical Software, 020201 artificial intelligence & image processing, business, Software
Abstract

International audience; The lattice Boltzmann method (LBM) is an innovative and promising approach in computational fluid dynamics. From an algorithmic standpoint it reduces to a regular data parallel procedure and is therefore well-suited to high performance computations. Numerous works report efficient implementations of the LBM for the GPU, but very few mention multi-GPU versions and even fewer GPU cluster implementations. Yet, to be of practical interest, GPU LBM solvers need to be able to perform large scale simulations. In the present contribution, we describe an efficient LBM implementation for CUDA GPU clusters. Our solver consists of a set of MPI communication routines and a CUDA kernel specifically designed to handle three-dimensional partitioning of the computation domain. Performance measurement were carried out on a small cluster. We show that the results are satisfying, both in terms of data throughput and parallelisation efficiency.

Published
2013
Full Text
View/download PDF

35. Towards Urban-Scale Flow Simulations Using the Lattice Boltzmann Method

Author

Christian Obrecht, Kuznik, F., Tourancheau, B., Roux, J. -J, Centre de Thermique de Lyon (CETHIL), Centre National de la Recherche Scientifique (CNRS)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA), Smart Wireless Networking (SWING), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-CITI Centre of Innovation in Telecommunications and Integration of services (CITI), Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA), Drakkar, Laboratoire d'Informatique de Grenoble (LIG), Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP)-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP)-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS), and Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)

Subjects
[INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI]
Abstract

Session: Validation, Calibration and Testing V - Publisher: www.ibpsa.org - Conference: www.bs2011.org; International audience; The lattice Boltzmann method (LBM) is an innovative approach in computational fluid dynamics (CFD). Due to the underlying lattice structure, the LBM is inherently parallel and therefore well suited for high performance computing. Emerging many-core devices, such as graphic processing units (GPUs), nowadays allow to run very large scale simulations on rather inexpensive hardware. In this contribution, we present some simulation results obtained using our multi-GPU LBM solver. For validation purpose, we study the flow around a wall-mounted cube and show good agreement with previously published results. Furthermore, we discuss larger scale flow simulations involving nine cubes which demonstrate the practicability of CFD simulations in building aeraulics.

Published
2011

36. The TheLMA project: Multi-GPU Implementation of the Lattice Boltzmann Method

Author

Bernard Tourancheau, Frédéric Kuznik, Christian Obrecht, Jean-Jacques Roux, Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), Laboratoire de l'Informatique du Parallélisme (LIP), Centre National de la Recherche Scientifique (CNRS)-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-École normale supérieure - Lyon (ENS Lyon), Laboratoire d'Informatique de Grenoble (LIG), Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)-Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF), École normale supérieure de Lyon (ENS de Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), and Université Pierre Mendès France - Grenoble 2 (UPMF)-Université Joseph Fourier - Grenoble 1 (UJF)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP )-Institut National Polytechnique de Grenoble (INPG)-Centre National de la Recherche Scientifique (CNRS)

Subjects
Speedup, Computer science, business.industry, Lattice Boltzmann methods, 010103 numerical & computational mathematics, GPU cluster, Parallel computing, Solver, Computational fluid dynamics, 01 natural sciences, 010305 fluids & plasmas, Theoretical Computer Science, Computational science, CUDA, Hardware and Architecture, 0103 physical sciences, Scalability, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], [PHYS.MECA.THER]Physics [physics]/Mechanics [physics]/Thermics [physics.class-ph], 0101 mathematics, General-purpose computing on graphics processing units, business, Software
Abstract

International audience; In this paper, we describe the implementation of a multi-graphical processing unit (GPU) fluid flow solver based on the lattice Boltzmann method (LBM). The LBM is a novel approach in computational fluid dynamics, with numerous interesting features from a computational, numerical, and physical standpoint. Our program is based on CUDA and uses POSIX threads to manage multiple computation devices. Using recently released hardware, our solver may therefore run eight GPUs in parallel, which allows us to perform simulations at a rather large scale. Performance and scalability are excellent, the speedup over sequential implementations being at least of two orders of magnitude. In addition, we discuss tiling and communication issues for present and forthcoming implementations.

Published
2011
Full Text
View/download PDF

37. Global Memory Access Modelling for Efficient Implementation of the Lattice Boltzmann Method on Graphics Processing Units

Author

Christian Obrecht, Bernard Tourancheau, Jean-Jacques Roux, Frédéric Kuznik, Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), Smart Wireless Networking (SWING), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-CITI Centre of Innovation in Telecommunications and Integration of services (CITI), Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA), Laboratoire de l'Informatique du Parallélisme (LIP), École normale supérieure de Lyon (ENS de Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), Algorithms and Scheduling for Distributed Heterogeneous Platforms (GRAAL), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de l'Informatique du Parallélisme (LIP), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure de Lyon (ENS de Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS), EDF, J.M.L.M. Palma et al., École normale supérieure - Lyon (ENS Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Lyon (ENS Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), and Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)

Subjects
[PHYS.PHYS.PHYS-FLU-DYN]Physics [physics]/Physics [physics]/Fluid Dynamics [physics.flu-dyn], Computer science, CUDA, Parallel computing, 01 natural sciences, Computational science, Scheduling (computing), [INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI], Approximation error, CUDA Pinned memory, 0103 physical sciences, 0101 mathematics, Graphics, 010306 general physics, ACM: J.: Computer Applications/J.2: PHYSICAL SCIENCES AND ENGINEERING, Uniform memory access, GPU computing, ACM: C.: Computer Systems Organization/C.2: COMPUTER-COMMUNICATION NETWORKS, 010101 applied mathematics, ACM: J.: Computer Applications, ACM: C.: Computer Systems Organization/C.4: PERFORMANCE OF SYSTEMS/C.4.3: Modeling techniques, lattice Boltzmann method, General-purpose computing on graphics processing units, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], CFD
Abstract

International audience; In this work, we investigate the global memory access mech- anism on recent GPUs. For the purpose of this study, we created spe- cific benchmark programs, which allowed us to explore the scheduling of global memory transactions. Thus, we formulate a model capable of estimating the execution time for a large class of applications. Our main goal is to facilitate optimisation of regular data-parallel applications on GPUs. As an example, we finally describe our CUDA implementations of LBM flow solvers on which our model was able to estimate performance with less than 5% relative error.

Published
2011

38. A new approach to the lattice Boltzmann method for graphics processing units

Author

Christian Obrecht, Jean-Jacques Roux, Frédéric Kuznik, Bernard Tourancheau, Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA), Smart Wireless Networking (SWING), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-CITI Centre of Innovation in Telecommunications and Integration of services (CITI), Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA), Laboratoire de l'Informatique du Parallélisme (LIP), École normale supérieure - Lyon (ENS Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), Algorithms and Scheduling for Distributed Heterogeneous Platforms (GRAAL), Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Laboratoire de l'Informatique du Parallélisme (LIP), Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Lyon (ENS Lyon)-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS), Université de Lyon-Université de Lyon-Institut National des Sciences Appliquées de Lyon (INSA Lyon), Université de Lyon-Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-Centre National de la Recherche Scientifique (CNRS), École normale supérieure de Lyon (ENS de Lyon)-Université Claude Bernard Lyon 1 (UCBL), and Université de Lyon-Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure de Lyon (ENS de Lyon)-Université Claude Bernard Lyon 1 (UCBL)

Subjects
Parallel computing, Computer science, Lattice Boltzmann method, Parallel algorithm, Lattice Boltzmann methods, CUDA, GPU programming, 01 natural sciences, 010305 fluids & plasmas, [INFO.INFO-NI]Computer Science [cs]/Networking and Internet Architecture [cs.NI], CUDA Pinned memory, Modelling and Simulation, 0103 physical sciences, 0101 mathematics, Graphics, 010101 applied mathematics, Computational Mathematics, Shared memory, Computational Theory and Mathematics, Modeling and Simulation, [PHYS.MECA.THER]Physics [physics]/Mechanics [physics]/Thermics [physics.class-ph], [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], General-purpose computing on graphics processing units, [INFO.INFO-DC]Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC], Data transmission
Abstract

International audience; Emerging many-core processors, like CUDA capable nVidia GPUs, are promising platforms for regular parallel algorithms such as the Lattice Boltzmann Method (LBM). Since the global memory for graphic devices shows high latency and LBM is data intensive, the memory access pattern is an important issue for achieving good performances. Whenever possible, global memory loads and stores should be coalescent and aligned, but the propagation phase in LBM can lead to frequent misaligned memory accesses. Most previous CUDA implementations of 3D LBM addressed this problem by using low latency on chip shared memory. Instead of this, our CUDA implementation of LBM follows carefully chosen data transfer schemes in global memory. For the 3D lid-driven cavity test case, we obtained up to 86% of the global memory maximal throughput on nVidia's GT200. We show that as a consequence highly efficient implementations of LBM on GPUs are possible, even for complex models.

Published
2010
Full Text
View/download PDF

39. LBM Based Flow Simulation Using GPU Computing Processor

Author

Gilles Rusaouen, Jean-Jacques Roux, Christian Obrecht, Frédéric Kuznik, Centre de Thermique de Lyon (CETHIL), Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS)-Institut National des Sciences Appliquées de Lyon (INSA Lyon), and Institut National des Sciences Appliquées (INSA)-Université de Lyon-Institut National des Sciences Appliquées (INSA)

Subjects
Floating point, Computer science, Lattice Boltzmann method, Lattice Boltzmann methods, Graphics processing unit, Double-precision floating-point format, Parallel computing, 01 natural sciences, Single-precision floating-point format, 010305 fluids & plasmas, Computational science, CUDA, Modelling and Simulation, 0103 physical sciences, 0101 mathematics, Graphics, ComputingMethodologies_COMPUTERGRAPHICS, 010101 applied mathematics, Computer Science::Performance, Computational Mathematics, Computational Theory and Mathematics, Fluid flow, Multi-threaded architecture, Modeling and Simulation, Computer Science::Mathematical Software, [SPI.MECA.THER]Engineering Sciences [physics]/Mechanics [physics.med-ph]/Thermics [physics.class-ph], [PHYS.MECA.THER]Physics [physics]/Mechanics [physics]/Thermics [physics.class-ph], General-purpose computing on graphics processing units, Graphics Processing Unit
Abstract

International audience; Graphics Processing Units (GPUs), originally developed for computer games, now provide computational power for scientific applications. In this paper, we develop a general purpose Lattice Boltzmann code that runs entirely on a single GPU. The results show that: (1) simple precision floating point arithmetic is sufficient for LBM computation in comparison to double precision; (2) the implementation of LBM on GPUs allows us to achieve up to about one billion lattice update per second using single precision floating point; (3) GPUs provide an inexpensive alternative to large clusters for fluid dynamics prediction.

Published
2010
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results