A Multilevel Parallelism Approach with MPI and OpenACC for Complex CFD Codes

McCall, Andrew J.; Roy, Christopher J.

doi:10.2514/6.2017-3293

Cited by 5 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These restrictions mean the following features cannot be used. As can be seen in [50] and [51], the 1st and 3rd restrictions do not have adverse effects on the performance. The 2nd restriction can be easily resolved by using Fortran subroutines instead of functions.…”

Section: V0: Baselinementioning

confidence: 93%

“…The baseline GPU version of SENSEI was implemented by McCall [50], [51]. McCall pointed out that there are some restrictions of the PGI compiler.…”

Section: V0: Baselinementioning

confidence: 99%

“…However, the temporary arrays deteriorate the code performance significantly. More details about these restrictions can be found in Ref [50] and [51].…”

Section: V0: Baselinementioning

confidence: 99%

“…Although the work in Ref [50] and [51] overcame many restrictions to port the code to the GPU, the GPU performance was not satisfactory. A NVIDIA P100 GPU was only 1.3x∼3.4x faster than a single Intel Xeon E5-2680v4 CPU core.…”

Section: V0: Baselinementioning

confidence: 99%

See 3 more Smart Citations

An Improved Framework of GPU Computing for CFD Applications on Structured Grids using OpenACC

Xue,

Jackson,

Roy

2020

Preprint

View full text Add to dashboard Cite

This paper is focused on improving multi-GPU performance of a research CFD code on structured grids. MPI and OpenACC directives are used to scale the code up to 16 GPUs. This paper shows that using 16 P100 GPUs and 16 V100 GPUs can be 30× and 70× faster than 16 Xeon CPU E5-2680v4 cores for three different test cases, respectively. A series of performance issues related to the scaling for the multi-block CFD code are addressed by applying various optimizations. Performance optimizations such as the pack/unpack message method, removing temporary arrays as arguments to procedure calls, allocating global memory for limiters and connected boundary data, reordering non-blocking MPI I send/I recv and Wait calls, reducing unnecessary implicit derived type member data movement between the host and the device and the use of GPUDirect can improve the compute utilization, memory throughput, and asynchronous progression in the multi-block CFD code using modern programming features.

show abstract

Section: V0: Baselinementioning

confidence: 93%

“…The baseline GPU version of SENSEI was implemented by McCall [50], [51]. McCall pointed out that there are some restrictions of the PGI compiler.…”

Section: V0: Baselinementioning

confidence: 99%

“…However, the temporary arrays deteriorate the code performance significantly. More details about these restrictions can be found in Ref [50] and [51].…”

Section: V0: Baselinementioning

confidence: 99%

Section: V0: Baselinementioning

confidence: 99%

See 2 more Smart Citations

An Improved Framework of GPU Computing for CFD Applications on Structured Grids using OpenACC

Xue,

Jackson,

Roy

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Use of OpenACC to parallelize CFD algorithms is described in [22]. Various other work has been done using OpenACC for CFD codes [19]. The focus of the current work differs from previous efforts in that we targeted multi-zone CFD codes implemented in Fortran and we are using OpenMP 4.5 as the API.…”

Section: Related Workmentioning

confidence: 99%

Enabling Execution of a Legacy CFD Mini Application on Accelerators Using OpenMP

Nompelis

Jost

Koniges

et al. 2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

We describe the process and outcome of our efforts to port a legacy Fortran benchmark code to heterogeneous GPU-accelerated computing architectures using OpenMP. The benchmark code is one of the multi-zone NAS Parallel Benchmarks (NPB-MZ) called SP-MZ. This "mini-app" mimics the computation and data movement that is found in popular legacy and modern implicit computational fluid dynamics (CFD) solvers. Our objective was to examine how efficiently legacy Fortran codes can be ported to accelerators by leveraging OpenMP directives. We describe the development and optimization process and demonstrate the performance impact of various code modifications. We show select profiling results from the NVIDIA Visual Profiler (nvpp) to help others diagnose and overcome performance issues in their own applications. We present results for two compute systems endowed with NVIDIA V100 accelerators.

show abstract

Optimization of Asynchronous Logging Kernels for a GPU Accelerated CFD Solver

Zehner¹,

Hashimoto²

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

A Multilevel Parallelism Approach with MPI and OpenACC for Complex CFD Codes

Cited by 5 publications

References 19 publications

An Improved Framework of GPU Computing for CFD Applications on Structured Grids using OpenACC

An Improved Framework of GPU Computing for CFD Applications on Structured Grids using OpenACC

Enabling Execution of a Legacy CFD Mini Application on Accelerators Using OpenMP

Optimization of Asynchronous Logging Kernels for a GPU Accelerated CFD Solver

Contact Info

Product

Resources

About