GPU-Accelerated simulation of elastic wave propagation

GPU-Accelerated simulation of elastic wave propagation

Stať ve sborníku v databázi WoS či Scopus

Modeling of ultrasound waves propagation in hard biological materials such as bones and skull has a rapidly growing area of applications, e.g. brain cancer treatment planing, deep brain neurostimulation and neuromodulation, and opening blood brain barriers. Recently, we have developed a novel numerical model of elastic wave propagation based on the Kelvin-Voigt model accounting for linear elastic wave proration in heterogeneous absorption media. Although, the model offers unprecedented fidelity, its computational requirements have been prohibitive for realistic simulations. This paper presents an optimized version of the simulation model accelerated by the Nvidia CUDA language and deployed on the best GPUs including the Nvidia P100 accelerators present in the Piz Daint supercomputer. The native CUDA code reaches a speed-up of 5.4 when compared to the Matlab prototype accelerated by the Parallel Computing Toolbox running on the same GPU. Such reduction in computation time enables computation of large-scale treatment plans in terms of hours.

Modeling of ultrasound waves propagation in hard biological materials such as bones and skull has a rapidly growing area of applications, e.g. brain cancer treatment planing, deep brain neurostimulation and neuromodulation, and opening blood brain barriers. Recently, we have developed a novel numerical model of elastic wave propagation based on the Kelvin-Voigt model accounting for linear elastic wave proration in heterogeneous absorption media. Although, the model offers unprecedented fidelity, its computational requirements have been prohibitive for realistic simulations. This paper presents an optimized version of the simulation model accelerated by the Nvidia CUDA language and deployed on the best GPUs including the Nvidia P100 accelerators present in the Piz Daint supercomputer. The native CUDA code reaches a speed-up of 5.4 when compared to the Matlab prototype accelerated by the Parallel Computing Toolbox running on the same GPU. Such reduction in computation time enables computation of large-scale treatment plans in terms of hours.

Ultrasound simulations, Elastic model, Pseudospectral methods, k-Wave toolbox, GPU

Ultrasound simulations, Elastic model, Pseudospectral methods, k-Wave toolbox, GPU

KADLUBIAK, K.; JAROŠ, J.; TREEBY, B.

2020

29.10.2018

IEEE Computer Society

Orleans

978-1-5386-7878-7

Proceedings - 2018 International Conference on High Performance Computing and Simulation, HPCS 2018

188

195

8

https://ieeexplore.ieee.org/document/8514349

HPCS 2018