TY - JOUR
T1 - GPU acceleration of CaNS for massively-parallel direct numerical simulations of canonical fluid flows
AU - Costa, Pedro
AU - Phillips, Everett
AU - Brandt, Luca
AU - Fatica, Massimiliano
PY - 2021
Y1 - 2021
N2 - This work presents the GPU acceleration of the open-source code CaNS for very fast massively-parallel simulations of canonical fluid flows. The distinct feature of the many-CPU Navier–Stokes solver in CaNS is its fast direct solver for the second-order finite-difference Poisson equation, based on the method of eigenfunction expansions. The solver implements all the boundary conditions valid for this type of problems in a unified framework. Here, we extend the solver for GPU-accelerated clusters using CUDA Fortran. The porting makes extensive use of CUF kernels and has been greatly simplified by the unified memory feature of CUDA Fortran, which handles the data migration between host (CPU) and device (GPU) without defining new arrays in the source code. The overall implementation has been validated against benchmark data for turbulent channel flow and its performance assessed on a NVIDIA DGX-2 system (16 T V100 32Gb, connected with NVLink via NVSwitch). The wall-clock time per time step of the GPU-accelerated implementation is impressively small when compared to its CPU implementation on state-of-the-art many-CPU clusters, as long as the domain partitioning is sufficiently small that the data resides mostly on the GPUs. The implementation has been made freely available and open source under the terms of an MIT license.
AB - This work presents the GPU acceleration of the open-source code CaNS for very fast massively-parallel simulations of canonical fluid flows. The distinct feature of the many-CPU Navier–Stokes solver in CaNS is its fast direct solver for the second-order finite-difference Poisson equation, based on the method of eigenfunction expansions. The solver implements all the boundary conditions valid for this type of problems in a unified framework. Here, we extend the solver for GPU-accelerated clusters using CUDA Fortran. The porting makes extensive use of CUF kernels and has been greatly simplified by the unified memory feature of CUDA Fortran, which handles the data migration between host (CPU) and device (GPU) without defining new arrays in the source code. The overall implementation has been validated against benchmark data for turbulent channel flow and its performance assessed on a NVIDIA DGX-2 system (16 T V100 32Gb, connected with NVLink via NVSwitch). The wall-clock time per time step of the GPU-accelerated implementation is impressively small when compared to its CPU implementation on state-of-the-art many-CPU clusters, as long as the domain partitioning is sufficiently small that the data resides mostly on the GPUs. The implementation has been made freely available and open source under the terms of an MIT license.
KW - Computational fluid dynamics
KW - Direct numerical simulation
KW - Fast Poisson solver
KW - GPU acceleration
UR - http://www.scopus.com/inward/record.url?scp=85078049969&partnerID=8YFLogxK
U2 - 10.1016/j.camwa.2020.01.002
DO - 10.1016/j.camwa.2020.01.002
M3 - Article
AN - SCOPUS:85078049969
VL - 81
SP - 502
EP - 511
JO - Computers & Mathematics with Applications
JF - Computers & Mathematics with Applications
SN - 0898-1221
ER -