Implicit method for the solution of supersonic and hypersonic 3D flow problems with Lower-Upper Symmetric-Gauss-Seidel preconditioner on multiple graphics processing units

Bocharov, A. N.; Евстигнеев, Н. М.; Petrovskiy, V P; Ryabkov, O. I.; Teplyakov, I. O.

doi:10.1016/j.jcp.2019.109189

Cited by 22 publications

(5 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Majority of simulation codes use CUDA, a proprietary framework only for NVIDIA GPUs. For instance, simulating compressible flows on multiple GPUs using an implicit method with a preconditioned BiCGSTAB solver is shown in [3], which is very close to what is being done in the present work. An example of large multi-GPU simulation of aerodynamic problems can be found in [4].…”

Section: Introductionsupporting

confidence: 84%

Adapting Complex and Clumsy CFD Code to Rapidly Changing Supercomputing Realities

Gorobets¹

2021

14th WCCM-ECCOMAS Congress

View full text Add to dashboard Cite

This work is devoted to acceleration and upgrade of the CFD code NOISEtte for scale-resolving simulations of compressible turbulent flows using edge-based high-accuracy methods on unstructured hybrid meshes. Attempts to extend the baseline multilevel MPI+OpenMP parallelization towards GPU-based hybrid systems have faced the problem: the code is too complex. It is an in-house research code with plenty of numerical methods, schemes, models, most of which are experimental and are not used in practical simulations. This chaotic zoo leads to excessive conditional branches, switches, redundant functional calls that slow down computations. Although the parallel algorithm is fully adapted to the stream processing paradigm, such an immense amount of code is too difficult to port efficiently to OpenCL or CUDA and maintain it in consistency with the CPU version. An approach to survive in the process of adaptation to hybrid systems has been elaborated. It consists of various components, such as creation of a simplified configurations, combining different stages of the algorithm in order to reduce memory traffic, collapsing multiple functions in one function without branches and switches, mixing single and double precision, etc. As a result, the upgraded code is about twice as fast on CPUs and can use GPUs from different manufacturers-AMD, NVIDIA, Intel through the OpenCL standard.

show abstract

Section: Introductionsupporting

confidence: 84%

Adapting Complex and Clumsy CFD Code to Rapidly Changing Supercomputing Realities

Gorobets¹

2021

14th WCCM-ECCOMAS Congress

View full text Add to dashboard Cite

show abstract

“…While for structured collocated grids quite easy reordering techniques are available, for instance the chess board colouring [73], performing an accurate and efficient reordering of elements in the case of unstructured grids can become quite challenging and in general requires the reordering of graphs, based on the local flow velocity and mesh connectivity, see e.g. [15,32,96].…”

Section: Computation Of W *mentioning

confidence: 99%

An Implicit Staggered Hybrid Finite Volume/Finite Element Solver for the Incompressible Navier-Stokes Equations

Alessia Lucca,

Saray Busto,

Michael Dumbser

2023

EAJAM

View full text Add to dashboard Cite

We present a novel fully implicit hybrid finite volume/finite element method for incompressible flows. Following previous works on semi-implicit hybrid FV/FE schemes, the incompressible Navier-Stokes equations are split into a pressure and a transportdiffusion subsystem. The first of them can be seen as a Poisson type problem and is thus solved efficiently using classical continuous Lagrange finite elements. On the other hand, finite volume methods are employed to solve the convective subsystem, in combination with Crouzeix-Raviart finite elements for the discretization of the viscous stress tensor. For some applications, the related CFL condition, even if depending only in the bulk velocity, may yield a severe time restriction in case explicit schemes are used. To overcome this issue an implicit approach is proposed. The system obtained from the implicit discretization of the transport-diffusion operator is solved using an inexact Newton-Krylov method, based either on the BiCStab or the GMRES algorithm. To improve the convergence properties of the linear solver a symmetric Gauss-Seidel (SGS) preconditioner is employed, together with a simple but efficient approach for the reordering of the grid elements that is compatible with MPI parallelization. Besides, considering the Ducros flux for the nonlinear convective terms we can prove that the discrete advection scheme is kinetic energy stable. The methodology is carefully assessed through a set of classical benchmarks for fluid mechanics. A last test shows the potential applicability of the method in the context of blood flow simulation in realistic vessel geometries.

show abstract

“…Many successful examples can be found in both scientific and commercial codes, see e.g. [1][2][3]8] among many others, or specifications of commercial codes capable of multi-GPU computing, such as Simcenter STAR-CCM+, Ansys Fluent multi-GPU solver, GPU-optimized Altair software, including AcuSolve or its LBM-based flow solver ultraFluidX [6].…”

Section: Introductionmentioning

confidence: 99%

Adapting a Scientific CFD Code to Industrial Applications on Hybrid Supercomputers

2022

JSFI

View full text Add to dashboard Cite

The NOISEtte heterogeneous parallel code for simulating turbulent flow and aerodynamic noise is considered. In our previous works, high acceleration and parallel efficiency in scientific scale-resolving simulations using GPUs were reported. For parallelization, the MPI, OpenMP and OpenCL standards are used, the latter allows using GPUs from different vendors. However, the further transition to industrial-oriented applications brought more trouble. Instead of discussing the parallel algorithm, this work will focus on the problems that are not so obvious at first glance, which arise when developing a heterogeneous simulation code. How to deal with numerous simulation algorithm components, all those bells and whistles like wall functions, mixing plane and sliding interfaces, synthetic turbulence generators, a variety of boundary conditions, etc., that either need to be ported to the GPU side or incorporated directly from the CPU side? How to maintain and modify the OpenCL code in a growing number of source files? How to arrange the modularity of a complex heterogeneous software package? How to preserve reliability and fault tolerance, especially in the case of numerical schemes of increased accuracy, but reduced social responsibility? These issues are discussed here and some solutions will be proposed.

show abstract

Implicit method for the solution of supersonic and hypersonic 3D flow problems with Lower-Upper Symmetric-Gauss-Seidel preconditioner on multiple graphics processing units

Cited by 22 publications

References 22 publications

Adapting Complex and Clumsy CFD Code to Rapidly Changing Supercomputing Realities

Adapting Complex and Clumsy CFD Code to Rapidly Changing Supercomputing Realities

An Implicit Staggered Hybrid Finite Volume/Finite Element Solver for the Incompressible Navier-Stokes Equations

Adapting a Scientific CFD Code to Industrial Applications on Hybrid Supercomputers

Contact Info

Product

Resources

About