Performance and Accuracy Analysis of Nonlinear k-Wave Simulations Using Local Domain Decomposition with an 8-GPU Server

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F18%3APU130766" target="_blank" >RIV/00216305:26230/18:PU130766 - isvavai.cz</a>

Výsledek na webu

<a href="https://asa.scitation.org/doi/10.1121/2.0000883" target="_blank" >https://asa.scitation.org/doi/10.1121/2.0000883</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1121/2.0000883" target="_blank" >10.1121/2.0000883</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Performance and Accuracy Analysis of Nonlinear k-Wave Simulations Using Local Domain Decomposition with an 8-GPU Server

Popis výsledku v původním jazyce

Large-scale nonlinear ultrasound simulations using the open-source k-Wave toolbox are now routinely performed using the MPI version of k-Wave running on traditional CPU-based clusters. However, the allto-all communications required by the 3D fast Fourier transform (FFT) severely impact performance when scaling to large numbers of compute cores. This can be overcome by using a domain decomposition strategy based on a local Fourier basis. In this work, we analyse the performance and accuracy of using local domain decomposition for running a high-intensity focused ultrasound (HIFU) simulation in the kidney on a single server containing eight NVIDIA P40 graphical processing units (GPUs). Different decompositions and overlap sizes are investigated and compared to a global MPI simulation running on a CPU-based supercomputer using 1280 cores. For a grid size of 960 × 960 × 1280 grid points and an overlap size of 4 grid points, the error in the simulation using local domain decomposition is on the order of 0.1% compared to the global simulation, which is sufficient for most applications. The financial cost for running the simulation is also reduced by more than an order of magnitude.

Název v anglickém jazyce

Performance and Accuracy Analysis of Nonlinear k-Wave Simulations Using Local Domain Decomposition with an 8-GPU Server

Popis výsledku anglicky

Large-scale nonlinear ultrasound simulations using the open-source k-Wave toolbox are now routinely performed using the MPI version of k-Wave running on traditional CPU-based clusters. However, the allto-all communications required by the 3D fast Fourier transform (FFT) severely impact performance when scaling to large numbers of compute cores. This can be overcome by using a domain decomposition strategy based on a local Fourier basis. In this work, we analyse the performance and accuracy of using local domain decomposition for running a high-intensity focused ultrasound (HIFU) simulation in the kidney on a single server containing eight NVIDIA P40 graphical processing units (GPUs). Different decompositions and overlap sizes are investigated and compared to a global MPI simulation running on a CPU-based supercomputer using 1280 cores. For a grid size of 960 × 960 × 1280 grid points and an overlap size of 4 grid points, the error in the simulation using local domain decomposition is on the order of 0.1% compared to the global simulation, which is sufficient for most applications. The financial cost for running the simulation is also reduced by more than an order of magnitude.

Druh

J_SC - Článek v periodiku v databázi SCOPUS

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

R - Projekt Ramcoveho programu EK

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Proceedings of Meetings on Acoustics

ISSN

1939-800X

e-ISSN

—

Svazek periodika

34

Číslo periodika v rámci svazku

1

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

5

Strana od-do

1-5

Kód UT WoS článku

—

EID výsledku v databázi Scopus

2-s2.0-85064972728