Acceleration of a two-dimensional Euler flow solver using commodity graphics hardware

Brandvik, Tobias; Pullan, Graham

doi:10.1243/09544062jmes813ft

Cited by 58 publications

(38 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This is shown in Figure 8, while the benchmarking hardware details are summarized in section 5.4.2. A 40x speed-up has been reported previously by the present authors [1] when comparing the same implementations, but running on a slower CPU. The large speed-up can be explained by the memory fetches performed by the solver having good 2D spatial locality, thereby making efficient use of the GPU's texture cache.…”

Section: Performancesupporting

confidence: 67%

See 1 more Smart Citation

Acceleration of a 3D Euler Solver Using Commodity Graphics Hardware

Brandvik

Pullan

2008

46th AIAA Aerospace Sciences Meeting and Exhibit

144

View full text Add to dashboard Cite

The porting of two-and three-dimensional Euler solvers from a conventional CPU implementation to the novel target platform of the Graphics Processing Unit (GPU) is described. The motivation for such an effort is the impressive performance that GPUs offer: typically 10 times more floating point operations per second than a modern CPU, with over 100 processing cores and all at a very modest financial cost. Both codes were found to generate the same results on the GPU as the FORTRAN versions did on the CPU. The 2D solver ran up to 29 times quicker on the GPU than on the CPU; the 3D solver 16 times faster.

show abstract

Section: Performancesupporting

confidence: 67%

“…The solutions presented are physically sensible and speed-ups compared to an equivalent CPU code of 10-20 times are reported. The present authors have recently published results from a 2D Euler solver [1] with speed-ups of up to 40 times obtained.…”

Section: Introductionmentioning

confidence: 99%

Acceleration of a 3D Euler Solver Using Commodity Graphics Hardware

Brandvik

Pullan

2008

46th AIAA Aerospace Sciences Meeting and Exhibit

144

View full text Add to dashboard Cite

show abstract

“…This situation has begun to improve with the recent release of the OpenCL standard, although this is not yet supported by all vendors. As an example of what may be possible, Brandvik and Pullan [37] implemented an Euler solver on a graphics processing unit (GPU) and reported speed-ups of up to 40 compared to a conventional Intel CPU. It remains to be seen if these impressive speed-ups can be maintained for a cluster of hundreds or thousands of GPUs when the issues of inter-GPU communication will become important.…”

Section: Parallelisationmentioning

confidence: 99%

Computational fluid dynamics and virtual aeroengine modelling

Chew

Hills

2009

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

Use of large scale computational fluid dynamics (CFD) models in aeroengine design has grown rapidly in recent years as parallel computing hardware has become available. This has reached the point where research aimed at the development of CFD-based "virtual engine test cells" is underway, with considerable debate of the subject within the industrial and research communities. The present paper considers and illustrates the state-of-the art and prospects for advances in this field. Limitations to CFD model accuracy, the need for aero-thermo-mechanical analysis through an engine flight cycle, coupling of numerical solutions for solid and fluid domains, and timescales for capability development are considered. While the fidelity of large scale CFD models will remain limited by turbulence modelling and other issues for the foreseeable future, it is clear that use of multi-scale, multi-physics modelling in engine design will expand considerably. Development of user-friendly, versatile, efficient programs and systems for use in a massively parallel computing environment is considered a key issue.

show abstract

“…To list a few of the CUDAaccelerated CFD applications, Elsen et al 11 reported a 3D high-order FDM solver for large calculation on multi-block structured grids; Klöckner et al 16 developed a 3D unstructured high-order nodal DGM solver for the Maxwell's equations; Corrigan et al 10 proposed a 3D FVM solver for compressible inviscid flows on unstructured tetrahedral grids; Zimmerman et al 29 presented an SDM solver for the Navier-Stokes equations on unstructured hexahedral grids; and more as in the references. 3,4,12,7,21,23,13,19,8,14,1,9 However applying CUDA to a legacy CFD code is not likely an easy job since the developer has to define an explicit layout of the threads on the GPU (numbers of blocks, numbers of threads) for each kernel function. 15 So what if the CFD code designers have to meet specific investment requirements like (1) enable GPU computing for legacy CFD programs at a minimum extra cost in time and effort (usually a major concern for large-scale code development), (2) enable the GPU-accelerated programs running on different platforms (similar to the situation that the video game designers would like to make their products available across platforms)?…”

Section: Introductionmentioning

confidence: 99%

OpenACC-based GPU Acceleration of a 3-D Unstructured Discontinuous Galerkin Method

Xia

Luo

et al. 2014

52nd Aerospace Sciences Meeting

View full text Add to dashboard Cite

A GPU-accelerated discontinuous Galerkin (DG) method is presented for the solution of compressible flows on 3-D unstructured grids. The present work has employed two of the most attractive features in a new programming standard of parallel computing -OpenACC: 1) multi-platform/compiler support and 2) descriptive directive interface to upgrade a legacy CFD solver with the capacity of GPU computing, without significant extra cost in recoding, resulting in a highly portable and extensible GPU-accelerated code. In addition, a face renumbering/grouping scheme is proposed to overcome the "race condition" in facebased flux calculations that occurs on GPU vectorization. Performance of the developed double-precision solver is assessed for both simple and complex geometries. Speedup factors up to but not limited to 24× and 1.6× were achieved by comparing the measured computing time of the OpenACC program running on an NVIDIA Tesla K20c GPU to that of the equivalent MPI program running on one single core and full sixteen cores of an AMD Opteron-6128 CPU respectively, indicating a great potential to port more features of the underlying DG solver into the OpenACC framework.

show abstract

Acceleration of a two-dimensional Euler flow solver using commodity graphics hardware

Cited by 58 publications

References 9 publications

Acceleration of a 3D Euler Solver Using Commodity Graphics Hardware

Acceleration of a 3D Euler Solver Using Commodity Graphics Hardware

Computational fluid dynamics and virtual aeroengine modelling

OpenACC-based GPU Acceleration of a 3-D Unstructured Discontinuous Galerkin Method

Contact Info

Product

Resources

About