A massively parallel and memory-efficient FEM toolbox with a hybrid total FETI solver with accelerator support
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27230%2F19%3A10240450" target="_blank" >RIV/61989100:27230/19:10240450 - isvavai.cz</a>
Nalezeny alternativní kódy
RIV/61989100:27240/19:10240450 RIV/61989100:27740/19:10240450
Výsledek na webu
<a href="https://doi.org/10.1177/1094342018798452" target="_blank" >https://doi.org/10.1177/1094342018798452</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1177/1094342018798452" target="_blank" >10.1177/1094342018798452</a>
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
A massively parallel and memory-efficient FEM toolbox with a hybrid total FETI solver with accelerator support
Popis výsledku v původním jazyce
In this article, we present the ExaScale PaRallel finite element tearing and interconnecting SOlver (ESPRESO) finite element method (FEM) library, which includes an FEM toolbox with interfaces to professional and open-source simulation tools, and a massively parallel hybrid total finite element tearing and interconnecting (HTFETI) solver which can fully utilize the Oak Ridge Leadership Computing Facility Titan supercomputer and achieve superlinear scaling. This article presents several new techniques for finite element tearing and interconnecting (FETI) solvers designed for efficient utilization of supercomputers with a focus on (i) performance-we present a fivefold reduction of solver runtime for the Laplace equation by redesigning the FETI solver and offloading the key workload to the accelerator. We compare Intel Xeon Phi 7120p and Tesla K80 and P100 accelerators to Intel Xeon E5-2680v3 and Xeon Phi 7210 central processing units; and (ii) memory efficiency-we present two techniques which increase the efficiency of the HTFETI solver 1.8 times and push the limits of the largest possible problem ESPRESO that can solve from 124 to 223 billion unknowns for problems with unstructured meshes. Finally, we show that by dynamically tuning hardware parameters, we can reduce energy consumption by up to 33%. (C) The Author(s) 2018.
Název v anglickém jazyce
A massively parallel and memory-efficient FEM toolbox with a hybrid total FETI solver with accelerator support
Popis výsledku anglicky
In this article, we present the ExaScale PaRallel finite element tearing and interconnecting SOlver (ESPRESO) finite element method (FEM) library, which includes an FEM toolbox with interfaces to professional and open-source simulation tools, and a massively parallel hybrid total finite element tearing and interconnecting (HTFETI) solver which can fully utilize the Oak Ridge Leadership Computing Facility Titan supercomputer and achieve superlinear scaling. This article presents several new techniques for finite element tearing and interconnecting (FETI) solvers designed for efficient utilization of supercomputers with a focus on (i) performance-we present a fivefold reduction of solver runtime for the Laplace equation by redesigning the FETI solver and offloading the key workload to the accelerator. We compare Intel Xeon Phi 7120p and Tesla K80 and P100 accelerators to Intel Xeon E5-2680v3 and Xeon Phi 7210 central processing units; and (ii) memory efficiency-we present two techniques which increase the efficiency of the HTFETI solver 1.8 times and push the limits of the largest possible problem ESPRESO that can solve from 124 to 223 billion unknowns for problems with unstructured meshes. Finally, we show that by dynamically tuning hardware parameters, we can reduce energy consumption by up to 33%. (C) The Author(s) 2018.
Klasifikace
Druh
J<sub>imp</sub> - Článek v periodiku v databázi Web of Science
CEP obor
—
OECD FORD obor
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Návaznosti výsledku
Projekt
Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.
Návaznosti
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach
Ostatní
Rok uplatnění
2019
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Údaje specifické pro druh výsledku
Název periodika
International Journal of High Performance Computing Applications
ISSN
1094-3420
e-ISSN
—
Svazek periodika
33
Číslo periodika v rámci svazku
19.9.2018
Stát vydavatele periodika
GB - Spojené království Velké Británie a Severního Irska
Počet stran výsledku
18
Strana od-do
660-677
Kód UT WoS článku
000471881700007
EID výsledku v databázi Scopus
2-s2.0-85059519221