Evelyn Otero - Academia.edu (original) (raw)

Papers by Evelyn Otero

An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigri... more An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigrid smoother combined with a line-implicit method as an acceleration technique for Reynolds-Averaged Navier-Stokes (RANS) simulation on stretched meshes. The Computational Fluid Dynamics code concerned is Edge, an edge-based finite volume Navier-Stokes flow solver for structured and unstructured grids. The paper focuses on the investigation of the parameters related to our novel line-implicit LU-SGS solver for convergence acceleration on 3D RANS meshes. The LU-SGS parameters are defined as the Courant-Friedrichs-Lewy number, the Left Hand Side dissipation, and the convergence of iterative solution of the linear problem arising from the linearisation of the implicit scheme. The influence of these parameters on the overall convergence is presented and default values are defined for maximum convergence acceleration. The optimized settings are applied to 3D RANS computations for comparison with explicit and line-implicit Runge-Kutta smoothing. For most of the cases, a computing time acceleration of the order of 2 is found depending on the mesh type, namely the boundary layer and the magnitude of residual reduction.

Computational fluid dynamics (CFD) has become a significant tool routinely used in design and opt... more Computational fluid dynamics (CFD) has become a significant tool routinely used in design and optimization in aerospace industry. Typical flows may be characterized by high-speed and compressible flow features and, in many cases, by massive flow separation and unsteadiness. Accurate and efficient numerical solution of time-dependent problems is hence required, and the efficiency of standard dual-time stepping methods used for unsteady flows in many CFD codes has been found inadequate for large-scale industrial problems. This has motivated the present work, in which major effort is made to replace the explicit relaxation methods with implicit time integration schemes. The CFD flow solver considered in this work is Edge, a node-based solver for unstructured grids based on a dual, edge-based formulation. Edge is the Swedish national CFD tool for computing compressible flow, used at the Swedish aircraft manufacturer SAAB, and developed at FOI, lately in collaboration with external national and international partners. The work is initially devoted to the implementation of an implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) type of relaxation in Edge with the purpose to speed up the convergence to steady state. The convergence of LU-SGS has been firstly accelerated by basing the implicit operator on a flux splitting method of matrix dissipation type. An increase of the diagonal dominance of the system matrix was the principal motivation. Then the code has been optimized by means of performance tools Intel Vtune and CrayPAT, improving the run time. It was found that the ordering of the unknowns significantly influences the convergence. Thus, different ordering techniques have been investigated. Finding the optimal ordering method is a very hard problem and the results obtained are mostly illustrative. Finally, to improve convergence speed on the stretched computational grids used for boundary layers LU-SGS has been combined with the line-implicit method.

Edge is a flow solver for unstructured grids based on a dual grid and edge-based formulation. The... more Edge is a flow solver for unstructured grids based on a dual grid and edge-based formulation. The standard dual-time stepping methods for compressible unsteady flows are inadequate for large-scale industrial problems. This has motivated the present work, in which an implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) type of relaxation has been implemented in the code Edge with multigrid acceleration. Two different types of dissipation, a scalar and a matrix model, have been constructed which increase the diagonal dominance of the system matrix but not the numerical viscosity of the computed solution. A parametric study demonstrates convergence accelerations by a factor of three for inviscid transonic flows compared to explicit Runge-Kutta smoothing for multigrid acceleration.

International Journal of Computational Fluid Dynamics, Feb 7, 2015

This is the accepted version of a paper published in. This paper has been peer-reviewed but does ... more This is the accepted version of a paper published in. This paper has been peer-reviewed but does not include the final publisher proof-corrections or journal pagination.

Journal of Parallel and Distributed Computing, Oct 1, 2019

Abstract Due to its high performance and throughput capabilities, GPU-accelerated computing is be... more Abstract Due to its high performance and throughput capabilities, GPU-accelerated computing is becoming a popular technology in scientific computing, in particular using programming models such as CUDA and OpenACC. The main advantage with OpenACC is that it enables to simply port codes in their “original” form to GPU systems through compiler directives, thus allowing an incremental approach. An OpenACC implementation is applied to the CFD code Nek5000 for simulation of incompressible flows, based on the spectral-element method. The work follows up previous implementations and focuses now on the P N − P N − 2 method for the spatial discretization of the Navier–Stokes equations. Performance results of the ported code show a speed-up of up to 3.1 on multi-GPU for a polynomial order N > 11 .

Computational fluid dynamics (CFD) is a significant tool routinely used indesign and optimization... more Computational fluid dynamics (CFD) is a significant tool routinely used indesign and optimization in aerospace industry. Often cases with unsteadyflows must be computed, and the long compute times of standard methods hasmotivated the present work on new implicit methods to replace the standardexplicit schemes. The implementation and numerical experiments were donewith the Swedish national flow solver Edge, developed by FOI,universities, and collaboration partners.The work is concentrated on a Lower-Upper Symmetric Gauss-Seidel (LU-SGS)type of time stepping. For the very anisotropic grids needed forReynolds-Averaged Navier-Stokes (RANS) computations of turbulent boundary layers,LU-SGS is combined with a line-implicit technique. The inviscid flux Jacobians which contribute to the diagonalblocks of the system matrix are based on a flux splitting method with upwind type dissipation giving control over diagonal dominance and artificial dissipation.The method is controlled by several parameters, and comprehensivenumerical experiments were carried out to identify their influence andinteraction so that close to optimal values can be suggested. As an example,the optimal number of iterations carried out in a time-step increases with increased resolution of the computational grid.The numbering of the unknowns is important, and the numberings produced by mesh generators of Delaunay- and advancing front-type wereamong the best.The solver has been parallelized with the Message Passing Interface (MPI) for runs on multi-processor hardware,and its performance scales with the number of processors at least asefficiently as the explicit methods. The new method saves typicallybetween 50 and 80 percent of the runtime, depending on the case, andthe largest computations have reached 110M grid nodes. Theclassical multigrid acceleration for 3D RANS simulations was foundineffective in the cases tested in combination with the LU-SGS solverusing optimal parameters. Finally, preliminary time-accurate simulations for unsteady flows have shown promising results.

Computational fluid dynamics (CFD) has become a significant tool routinely used in design and opt... more Computational fluid dynamics (CFD) has become a significant tool routinely used in design and optimization in aerospace industry. Typical flows may be characterized by high-speed and compressible flow features and, in many cases, by massive flow separation and unsteadiness. Accurate and efficient numerical solution of time-dependent problems is hence required, and the efficiency of standard dual-time stepping methods used for unsteady flows in many CFD codes has been found inadequate for large-scale industrial problems. This has motivated the present work, in which major effort is made to replace the explicit relaxation methods with implicit time integration schemes. The CFD flow solver considered in this work is Edge, a node-based solver for unstructured grids based on a dual, edge-based formulation. Edge is the Swedish national CFD tool for computing compressible flow, used at the Swedish aircraft manufacturer SAAB, and developed at FOI, lately in collaboration with external national and international partners. The work is initially devoted to the implementation of an implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) type of relaxation in Edge with the purpose to speed up the convergence to steady state. The convergence of LU-SGS has been firstly accelerated by basing the implicit operator on a flux splitting method of matrix dissipation type. An increase of the diagonal dominance of the system matrix was the principal motivation. Then the code has been optimized by means of performance tools Intel Vtune and CrayPAT, improving the run time. It was found that the ordering of the unknowns significantly influences the convergence. Thus, different ordering techniques have been investigated. Finding the optimal ordering method is a very hard problem and the results obtained are mostly illustrative. Finally, to improve convergence speed on the stretched computational grids used for boundary layers LU-SGS has been combined with the line-implicit method.

The implicit LU-SGS solver has been implemented in the code Edge to accelerate the convergence to... more The implicit LU-SGS solver has been implemented in the code Edge to accelerate the convergence to steady state. Edge is a flow solver for unstructured grids based on a dual grid and edgebased formulation. LU-SGS has been combined with the line-implicit technique to improve convergence on the very anisotropic grids necessary for the boundary layers. LU-SGS works in parallel and gives better linear scaling with respect to the number of processors, than the explicit scheme. The ordering techniques investigated have shown that node numbering does influence the convergence and that the native orderings from Delaunay and advancing front generation were among the best tested. LU-SGS for 2D Euler and line-implicit LU-SGS for 2D RANS are two to three times faster than the explicit and line-implicit Runge-Kutta respectively. 3D cases show less acceleration and need a deeper study.

Flow, turbulence and combustion, May 25, 2018

Postprocessing and storage of large data sets represent one of the main computational bottlenecks... more Postprocessing and storage of large data sets represent one of the main computational bottlenecks in computational fluid dynamics. We assume that the accuracy necessary for computation is higher than needed for postprocessing. Therefore, in the current work we assess thresholds for data reduction as required by the most common data analysis tools used in the study of fluid flow phenomena, specifically wall-bounded turbulence. These thresholds are imposed a priori by the user in L 2-norm, and we assess a set of parameters to identify the minimum accuracy requirements. The method considered in the present work is the discrete Legendre transform (DLT), which we evaluate in the computation of turbulence statistics, spectral analysis and resilience for cases highly-sensitive to the initial conditions. Maximum acceptable compression ratios of the original data have been found to be around 97%, depending on the application purpose. The new method outperforms downsampling, as well as the previously explored data truncation method based on discrete Chebyshev transform (DCT).

An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigri... more An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigrid (MG) smoother combined with a line-implicit method as an acceleration technique for Reynolds-Averaged Navier-Stokes (RANS) simulation on stretched meshes. The Computational Fluid Dynamics (CFD) code concerned is Edge, an edge- based finite volume Navier-Stokes flow solver for structured and unstructured grids. The paper focuses on the investigation of the parameters related to the novel line-implicit LU- SGS solver for convergence acceleration on 3D RANS meshes. The influence on the overall convergence of the Courant-Friedrichs-Lewy (CFL) number, the Left Hand Side (LHS) dissipation, and the convergence of iterative solution of the linear problem is presented and default values are defined for maximum convergence acceleration. These optimized set- tings are applied to 3D RANS computations for comparison with explicit and line-implicit Runge-Kutta (RK) smoothing. For most of the cases, a computing time acceleration of the order of 2 is found depending on the mesh type, namely the boundary layer and the magnitude of residual reduction.

ERCOFTAC series, 2019

The field of computational fluid dynamics (CFD) is data intensive, particularly for high-fidelity... more The field of computational fluid dynamics (CFD) is data intensive, particularly for high-fidelity simulations. Direct and large-eddy simulations (DNS and LES), which are framed in this high-fidelity regime, require to capture a wide range of flow scales, a fact that leads to a high number of degrees of freedom. Besides the computational bottleneck, brought by the size of the problem, a slightly overlooked issue is the manipulation of the data. High amounts of disk space and also the slow speed of I/O (input/output) impose limitations on large-scale simulations. Typically the computational requirements for proper resolution of the flow structures are far higher than those of post-processing. To mitigate such shortcomings we employ a lossy data compression procedure, and track the reduction that occurs for various levels of truncation of the data set.

The 5th International Conference on Exascale Applications and Software, 17th to 19th April 2018 in Edinburgh, Scotland, 2018

An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigri... more An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigrid (MG) smoother combined with a line-implicit method as an acceleration technique for Reynolds-Averaged Navier-Stokes (RANS) simulation on stretched meshes. The Computational Fluid Dynamics (CFD) code concerned is Edge, an edge- based finite volume Navier-Stokes flow solver for structured and unstructured grids. The paper focuses on the investigation of the parameters related to the novel line-implicit LU- SGS solver for convergence acceleration on 3D RANS meshes. The influence on the overall convergence of the Courant-Friedrichs-Lewy (CFL) number, the Left Hand Side (LHS) dissipation, and the convergence of iterative solution of the linear problem is presented and default values are defined for maximum convergence acceleration. These optimized set- tings are applied to 3D RANS computations for comparison with explicit and line-implicit Runge-Kutta (RK) smoothing. For most of the cases, ...

Direct and Large-Eddy Simulation XI, 2019

The field of computational fluid dynamics (CFD) is data intensive, particularly for high-fidelity... more The field of computational fluid dynamics (CFD) is data intensive, particularly for high-fidelity simulations. Direct and large-eddy simulations (DNS and LES), which are framed in this high-fidelity regime, require to capture a wide range of flow scales, a fact that leads to a high number of degrees of freedom. Besides the computational bottleneck, brought by the size of the problem, a slightly overlooked issue is the manipulation of the data. High amounts of disk space and also the slow speed of I/O (input/output) impose limitations on large-scale simulations. Typically the computational requirements for proper resolution of the flow structures are far higher than those of post-processing. To mitigate such shortcomings we employ a lossy data compression procedure, and track the reduction that occurs for various levels of truncation of the data set.

CEAS Aeronautical Journal, 2021

Traveling and possible impact on climate and environment are currently under intense debate, and ... more Traveling and possible impact on climate and environment are currently under intense debate, and air travel in particular is often in question due to the use of fossil fuels. Electric propulsion has therefore become very popular but the energy sources for electricity generation should as well be taken into consideration. On the other hand, the social aspect of traveling is usually forgotten and should be also included for a complete sustainability analysis. In this study, the business trip from Stockholm to Bordeaux experienced by airplane and train is analyzed. Though the journey by airplane generated six and a half times more CO2 emissions than the journey by train on a per-passenger basis, this latter resulted in a 35-h journey compared to seven, and a cost up to eight and a half times more expensive than the airplane. The trip is defined as an optimization problem with focus on environmental, economic, and social impact to define acceptable trade-offs. The critical criteria for ...

Flow, Turbulence and Combustion, 2018

Postprocessing and storage of large data sets represent one of the main computational bottlenecks... more Postprocessing and storage of large data sets represent one of the main computational bottlenecks in computational fluid dynamics. We assume that the accuracy necessary for computation is higher than needed for postprocessing. Therefore, in the current work we assess thresholds for data reduction as required by the most common data analysis tools used in the study of fluid flow phenomena, specifically wall-bounded turbulence. These thresholds are imposed a priori by the user in L 2-norm, and we assess a set of parameters to identify the minimum accuracy requirements. The method considered in the present work is the discrete Legendre transform (DLT), which we evaluate in the computation of turbulence statistics, spectral analysis and resilience for cases highly-sensitive to the initial conditions. Maximum acceptable compression ratios of the original data have been found to be around 97%, depending on the application purpose. The new method outperforms downsampling, as well as the previously explored data truncation method based on discrete Chebyshev transform (DCT).

Journal of Parallel and Distributed Computing, 2019

Abstract Due to its high performance and throughput capabilities, GPU-accelerated computing is be... more Abstract Due to its high performance and throughput capabilities, GPU-accelerated computing is becoming a popular technology in scientific computing, in particular using programming models such as CUDA and OpenACC. The main advantage with OpenACC is that it enables to simply port codes in their “original” form to GPU systems through compiler directives, thus allowing an incremental approach. An OpenACC implementation is applied to the CFD code Nek5000 for simulation of incompressible flows, based on the spectral-element method. The work follows up previous implementations and focuses now on the P N − P N − 2 method for the spatial discretization of the Navier–Stokes equations. Performance results of the ported code show a speed-up of up to 3.1 on multi-GPU for a polynomial order N > 11 .

Computational fluid dynamics (CFD) has become a significant tool routinely used in design and opt... more Computational fluid dynamics (CFD) has become a significant tool routinely used in design and optimization in aerospace industry. Typical flows may be characterized by high-speed and compressible flow features and, in many cases, by massive flow separation and unsteadiness. Accurate and efficient numerical solution of time-dependent problems is hence required, and the efficiency of standard dual-time stepping methods used for unsteady flows in many CFD codes has been found inadequate for large-scale industrial problems. This has motivated the present work, in which major effort is made to replace the explicit relaxation methods with implicit time integration schemes. The CFD flow solver considered in this work is Edge, a node-based solver for unstructured grids based on a dual, edge-based formulation. Edge is the Swedish national CFD tool for computing compressible flow, used at the Swedish aircraft manufacturer SAAB, and developed at FOI, lately in collaboration with external national and international partners. The work is initially devoted to the implementation of an implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) type of relaxation in Edge with the purpose to speed up the convergence to steady state. The convergence of LU-SGS has been firstly accelerated by basing the implicit operator on a flux splitting method of matrix dissipation type. An increase of the diagonal dominance of the system matrix was the principal motivation. Then the code has been optimized by means of performance tools Intel Vtune and CrayPAT, improving the run time. It was found that the ordering of the unknowns significantly influences the convergence. Thus, different ordering techniques have been investigated. Finding the optimal ordering method is a very hard problem and the results obtained are mostly illustrative. Finally, to improve convergence speed on the stretched computational grids used for boundary layers LU-SGS has been combined with the line-implicit method.

An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigri... more An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigrid smoother combined with a line-implicit method as an acceleration technique for Reynolds-Averaged Navier-Stokes (RANS) simulation on stretched meshes. The Computational Fluid Dynamics code concerned is Edge, an edge-based finite volume Navier-Stokes flow solver for structured and unstructured grids. The paper focuses on the investigation of the parameters related to our novel line-implicit LU-SGS solver for convergence acceleration on 3D RANS meshes. The LU-SGS parameters are defined as the Courant-Friedrichs-Lewy number, the Left Hand Side dissipation, and the convergence of iterative solution of the linear problem arising from the linearisation of the implicit scheme. The influence of these parameters on the overall convergence is presented and default values are defined for maximum convergence acceleration. The optimized settings are applied to 3D RANS computations for comparison with explicit and line-implicit Runge-Kutta smoothing. For most of the cases, a computing time acceleration of the order of 2 is found depending on the mesh type, namely the boundary layer and the magnitude of residual reduction.

Computational fluid dynamics (CFD) has become a significant tool routinely used in design and opt... more Computational fluid dynamics (CFD) has become a significant tool routinely used in design and optimization in aerospace industry. Typical flows may be characterized by high-speed and compressible flow features and, in many cases, by massive flow separation and unsteadiness. Accurate and efficient numerical solution of time-dependent problems is hence required, and the efficiency of standard dual-time stepping methods used for unsteady flows in many CFD codes has been found inadequate for large-scale industrial problems. This has motivated the present work, in which major effort is made to replace the explicit relaxation methods with implicit time integration schemes. The CFD flow solver considered in this work is Edge, a node-based solver for unstructured grids based on a dual, edge-based formulation. Edge is the Swedish national CFD tool for computing compressible flow, used at the Swedish aircraft manufacturer SAAB, and developed at FOI, lately in collaboration with external national and international partners. The work is initially devoted to the implementation of an implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) type of relaxation in Edge with the purpose to speed up the convergence to steady state. The convergence of LU-SGS has been firstly accelerated by basing the implicit operator on a flux splitting method of matrix dissipation type. An increase of the diagonal dominance of the system matrix was the principal motivation. Then the code has been optimized by means of performance tools Intel Vtune and CrayPAT, improving the run time. It was found that the ordering of the unknowns significantly influences the convergence. Thus, different ordering techniques have been investigated. Finding the optimal ordering method is a very hard problem and the results obtained are mostly illustrative. Finally, to improve convergence speed on the stretched computational grids used for boundary layers LU-SGS has been combined with the line-implicit method.

Edge is a flow solver for unstructured grids based on a dual grid and edge-based formulation. The... more Edge is a flow solver for unstructured grids based on a dual grid and edge-based formulation. The standard dual-time stepping methods for compressible unsteady flows are inadequate for large-scale industrial problems. This has motivated the present work, in which an implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) type of relaxation has been implemented in the code Edge with multigrid acceleration. Two different types of dissipation, a scalar and a matrix model, have been constructed which increase the diagonal dominance of the system matrix but not the numerical viscosity of the computed solution. A parametric study demonstrates convergence accelerations by a factor of three for inviscid transonic flows compared to explicit Runge-Kutta smoothing for multigrid acceleration.

International Journal of Computational Fluid Dynamics, Feb 7, 2015

This is the accepted version of a paper published in. This paper has been peer-reviewed but does ... more This is the accepted version of a paper published in. This paper has been peer-reviewed but does not include the final publisher proof-corrections or journal pagination.

Journal of Parallel and Distributed Computing, Oct 1, 2019

Abstract Due to its high performance and throughput capabilities, GPU-accelerated computing is be... more Abstract Due to its high performance and throughput capabilities, GPU-accelerated computing is becoming a popular technology in scientific computing, in particular using programming models such as CUDA and OpenACC. The main advantage with OpenACC is that it enables to simply port codes in their “original” form to GPU systems through compiler directives, thus allowing an incremental approach. An OpenACC implementation is applied to the CFD code Nek5000 for simulation of incompressible flows, based on the spectral-element method. The work follows up previous implementations and focuses now on the P N − P N − 2 method for the spatial discretization of the Navier–Stokes equations. Performance results of the ported code show a speed-up of up to 3.1 on multi-GPU for a polynomial order N > 11 .

Computational fluid dynamics (CFD) is a significant tool routinely used indesign and optimization... more Computational fluid dynamics (CFD) is a significant tool routinely used indesign and optimization in aerospace industry. Often cases with unsteadyflows must be computed, and the long compute times of standard methods hasmotivated the present work on new implicit methods to replace the standardexplicit schemes. The implementation and numerical experiments were donewith the Swedish national flow solver Edge, developed by FOI,universities, and collaboration partners.The work is concentrated on a Lower-Upper Symmetric Gauss-Seidel (LU-SGS)type of time stepping. For the very anisotropic grids needed forReynolds-Averaged Navier-Stokes (RANS) computations of turbulent boundary layers,LU-SGS is combined with a line-implicit technique. The inviscid flux Jacobians which contribute to the diagonalblocks of the system matrix are based on a flux splitting method with upwind type dissipation giving control over diagonal dominance and artificial dissipation.The method is controlled by several parameters, and comprehensivenumerical experiments were carried out to identify their influence andinteraction so that close to optimal values can be suggested. As an example,the optimal number of iterations carried out in a time-step increases with increased resolution of the computational grid.The numbering of the unknowns is important, and the numberings produced by mesh generators of Delaunay- and advancing front-type wereamong the best.The solver has been parallelized with the Message Passing Interface (MPI) for runs on multi-processor hardware,and its performance scales with the number of processors at least asefficiently as the explicit methods. The new method saves typicallybetween 50 and 80 percent of the runtime, depending on the case, andthe largest computations have reached 110M grid nodes. Theclassical multigrid acceleration for 3D RANS simulations was foundineffective in the cases tested in combination with the LU-SGS solverusing optimal parameters. Finally, preliminary time-accurate simulations for unsteady flows have shown promising results.

Computational fluid dynamics (CFD) has become a significant tool routinely used in design and opt... more Computational fluid dynamics (CFD) has become a significant tool routinely used in design and optimization in aerospace industry. Typical flows may be characterized by high-speed and compressible flow features and, in many cases, by massive flow separation and unsteadiness. Accurate and efficient numerical solution of time-dependent problems is hence required, and the efficiency of standard dual-time stepping methods used for unsteady flows in many CFD codes has been found inadequate for large-scale industrial problems. This has motivated the present work, in which major effort is made to replace the explicit relaxation methods with implicit time integration schemes. The CFD flow solver considered in this work is Edge, a node-based solver for unstructured grids based on a dual, edge-based formulation. Edge is the Swedish national CFD tool for computing compressible flow, used at the Swedish aircraft manufacturer SAAB, and developed at FOI, lately in collaboration with external national and international partners. The work is initially devoted to the implementation of an implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) type of relaxation in Edge with the purpose to speed up the convergence to steady state. The convergence of LU-SGS has been firstly accelerated by basing the implicit operator on a flux splitting method of matrix dissipation type. An increase of the diagonal dominance of the system matrix was the principal motivation. Then the code has been optimized by means of performance tools Intel Vtune and CrayPAT, improving the run time. It was found that the ordering of the unknowns significantly influences the convergence. Thus, different ordering techniques have been investigated. Finding the optimal ordering method is a very hard problem and the results obtained are mostly illustrative. Finally, to improve convergence speed on the stretched computational grids used for boundary layers LU-SGS has been combined with the line-implicit method.

The implicit LU-SGS solver has been implemented in the code Edge to accelerate the convergence to... more The implicit LU-SGS solver has been implemented in the code Edge to accelerate the convergence to steady state. Edge is a flow solver for unstructured grids based on a dual grid and edgebased formulation. LU-SGS has been combined with the line-implicit technique to improve convergence on the very anisotropic grids necessary for the boundary layers. LU-SGS works in parallel and gives better linear scaling with respect to the number of processors, than the explicit scheme. The ordering techniques investigated have shown that node numbering does influence the convergence and that the native orderings from Delaunay and advancing front generation were among the best tested. LU-SGS for 2D Euler and line-implicit LU-SGS for 2D RANS are two to three times faster than the explicit and line-implicit Runge-Kutta respectively. 3D cases show less acceleration and need a deeper study.

Flow, turbulence and combustion, May 25, 2018

Postprocessing and storage of large data sets represent one of the main computational bottlenecks... more Postprocessing and storage of large data sets represent one of the main computational bottlenecks in computational fluid dynamics. We assume that the accuracy necessary for computation is higher than needed for postprocessing. Therefore, in the current work we assess thresholds for data reduction as required by the most common data analysis tools used in the study of fluid flow phenomena, specifically wall-bounded turbulence. These thresholds are imposed a priori by the user in L 2-norm, and we assess a set of parameters to identify the minimum accuracy requirements. The method considered in the present work is the discrete Legendre transform (DLT), which we evaluate in the computation of turbulence statistics, spectral analysis and resilience for cases highly-sensitive to the initial conditions. Maximum acceptable compression ratios of the original data have been found to be around 97%, depending on the application purpose. The new method outperforms downsampling, as well as the previously explored data truncation method based on discrete Chebyshev transform (DCT).

An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigri... more An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigrid (MG) smoother combined with a line-implicit method as an acceleration technique for Reynolds-Averaged Navier-Stokes (RANS) simulation on stretched meshes. The Computational Fluid Dynamics (CFD) code concerned is Edge, an edge- based finite volume Navier-Stokes flow solver for structured and unstructured grids. The paper focuses on the investigation of the parameters related to the novel line-implicit LU- SGS solver for convergence acceleration on 3D RANS meshes. The influence on the overall convergence of the Courant-Friedrichs-Lewy (CFL) number, the Left Hand Side (LHS) dissipation, and the convergence of iterative solution of the linear problem is presented and default values are defined for maximum convergence acceleration. These optimized set- tings are applied to 3D RANS computations for comparison with explicit and line-implicit Runge-Kutta (RK) smoothing. For most of the cases, a computing time acceleration of the order of 2 is found depending on the mesh type, namely the boundary layer and the magnitude of residual reduction.

ERCOFTAC series, 2019

The field of computational fluid dynamics (CFD) is data intensive, particularly for high-fidelity... more The field of computational fluid dynamics (CFD) is data intensive, particularly for high-fidelity simulations. Direct and large-eddy simulations (DNS and LES), which are framed in this high-fidelity regime, require to capture a wide range of flow scales, a fact that leads to a high number of degrees of freedom. Besides the computational bottleneck, brought by the size of the problem, a slightly overlooked issue is the manipulation of the data. High amounts of disk space and also the slow speed of I/O (input/output) impose limitations on large-scale simulations. Typically the computational requirements for proper resolution of the flow structures are far higher than those of post-processing. To mitigate such shortcomings we employ a lossy data compression procedure, and track the reduction that occurs for various levels of truncation of the data set.

The 5th International Conference on Exascale Applications and Software, 17th to 19th April 2018 in Edinburgh, Scotland, 2018

An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigri... more An implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) solver has been implemented as a multigrid (MG) smoother combined with a line-implicit method as an acceleration technique for Reynolds-Averaged Navier-Stokes (RANS) simulation on stretched meshes. The Computational Fluid Dynamics (CFD) code concerned is Edge, an edge- based finite volume Navier-Stokes flow solver for structured and unstructured grids. The paper focuses on the investigation of the parameters related to the novel line-implicit LU- SGS solver for convergence acceleration on 3D RANS meshes. The influence on the overall convergence of the Courant-Friedrichs-Lewy (CFL) number, the Left Hand Side (LHS) dissipation, and the convergence of iterative solution of the linear problem is presented and default values are defined for maximum convergence acceleration. These optimized set- tings are applied to 3D RANS computations for comparison with explicit and line-implicit Runge-Kutta (RK) smoothing. For most of the cases, ...

Direct and Large-Eddy Simulation XI, 2019

The field of computational fluid dynamics (CFD) is data intensive, particularly for high-fidelity... more The field of computational fluid dynamics (CFD) is data intensive, particularly for high-fidelity simulations. Direct and large-eddy simulations (DNS and LES), which are framed in this high-fidelity regime, require to capture a wide range of flow scales, a fact that leads to a high number of degrees of freedom. Besides the computational bottleneck, brought by the size of the problem, a slightly overlooked issue is the manipulation of the data. High amounts of disk space and also the slow speed of I/O (input/output) impose limitations on large-scale simulations. Typically the computational requirements for proper resolution of the flow structures are far higher than those of post-processing. To mitigate such shortcomings we employ a lossy data compression procedure, and track the reduction that occurs for various levels of truncation of the data set.

CEAS Aeronautical Journal, 2021

Traveling and possible impact on climate and environment are currently under intense debate, and ... more Traveling and possible impact on climate and environment are currently under intense debate, and air travel in particular is often in question due to the use of fossil fuels. Electric propulsion has therefore become very popular but the energy sources for electricity generation should as well be taken into consideration. On the other hand, the social aspect of traveling is usually forgotten and should be also included for a complete sustainability analysis. In this study, the business trip from Stockholm to Bordeaux experienced by airplane and train is analyzed. Though the journey by airplane generated six and a half times more CO2 emissions than the journey by train on a per-passenger basis, this latter resulted in a 35-h journey compared to seven, and a cost up to eight and a half times more expensive than the airplane. The trip is defined as an optimization problem with focus on environmental, economic, and social impact to define acceptable trade-offs. The critical criteria for ...

Flow, Turbulence and Combustion, 2018

Postprocessing and storage of large data sets represent one of the main computational bottlenecks... more Postprocessing and storage of large data sets represent one of the main computational bottlenecks in computational fluid dynamics. We assume that the accuracy necessary for computation is higher than needed for postprocessing. Therefore, in the current work we assess thresholds for data reduction as required by the most common data analysis tools used in the study of fluid flow phenomena, specifically wall-bounded turbulence. These thresholds are imposed a priori by the user in L 2-norm, and we assess a set of parameters to identify the minimum accuracy requirements. The method considered in the present work is the discrete Legendre transform (DLT), which we evaluate in the computation of turbulence statistics, spectral analysis and resilience for cases highly-sensitive to the initial conditions. Maximum acceptable compression ratios of the original data have been found to be around 97%, depending on the application purpose. The new method outperforms downsampling, as well as the previously explored data truncation method based on discrete Chebyshev transform (DCT).

Journal of Parallel and Distributed Computing, 2019

Abstract Due to its high performance and throughput capabilities, GPU-accelerated computing is be... more Abstract Due to its high performance and throughput capabilities, GPU-accelerated computing is becoming a popular technology in scientific computing, in particular using programming models such as CUDA and OpenACC. The main advantage with OpenACC is that it enables to simply port codes in their “original” form to GPU systems through compiler directives, thus allowing an incremental approach. An OpenACC implementation is applied to the CFD code Nek5000 for simulation of incompressible flows, based on the spectral-element method. The work follows up previous implementations and focuses now on the P N − P N − 2 method for the spatial discretization of the Navier–Stokes equations. Performance results of the ported code show a speed-up of up to 3.1 on multi-GPU for a polynomial order N > 11 .

Computational fluid dynamics (CFD) has become a significant tool routinely used in design and opt... more Computational fluid dynamics (CFD) has become a significant tool routinely used in design and optimization in aerospace industry. Typical flows may be characterized by high-speed and compressible flow features and, in many cases, by massive flow separation and unsteadiness. Accurate and efficient numerical solution of time-dependent problems is hence required, and the efficiency of standard dual-time stepping methods used for unsteady flows in many CFD codes has been found inadequate for large-scale industrial problems. This has motivated the present work, in which major effort is made to replace the explicit relaxation methods with implicit time integration schemes. The CFD flow solver considered in this work is Edge, a node-based solver for unstructured grids based on a dual, edge-based formulation. Edge is the Swedish national CFD tool for computing compressible flow, used at the Swedish aircraft manufacturer SAAB, and developed at FOI, lately in collaboration with external national and international partners. The work is initially devoted to the implementation of an implicit Lower-Upper Symmetric Gauss-Seidel (LU-SGS) type of relaxation in Edge with the purpose to speed up the convergence to steady state. The convergence of LU-SGS has been firstly accelerated by basing the implicit operator on a flux splitting method of matrix dissipation type. An increase of the diagonal dominance of the system matrix was the principal motivation. Then the code has been optimized by means of performance tools Intel Vtune and CrayPAT, improving the run time. It was found that the ordering of the unknowns significantly influences the convergence. Thus, different ordering techniques have been investigated. Finding the optimal ordering method is a very hard problem and the results obtained are mostly illustrative. Finally, to improve convergence speed on the stretched computational grids used for boundary layers LU-SGS has been combined with the line-implicit method.