GPU-acceleration of the ELPA2 distributed eigensolver for dense symmetric and hermitian eigenproblems

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985840%3A_____%2F21%3A00539376" target="_blank" >RIV/67985840:_____/21:00539376 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1016/j.cpc.2020.107808" target="_blank" >https://doi.org/10.1016/j.cpc.2020.107808</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1016/j.cpc.2020.107808" target="_blank" >10.1016/j.cpc.2020.107808</a>

Jazyk výsledku

angličtina

Název v původním jazyce

GPU-acceleration of the ELPA2 distributed eigensolver for dense symmetric and hermitian eigenproblems

Popis výsledku v původním jazyce

The solution of eigenproblems is often a key computational bottleneck that limits the tractable system size of numerical algorithms, among them electronic structure theory in chemistry and in condensed matter physics. Large eigenproblems can easily exceed the capacity of a single compute node, thus must be solved on distributed-memory parallel computers. We here present GPU-oriented optimizations of the ELPA two-stage tridiagonalization eigensolver (ELPA2). On top of cuBLAS-based GPU offloading, we add a CUDA kernel to speed up the back-transformation of eigenvectors, which can be the computationally most expensive part of the two-stage tridiagonalization algorithm. We benchmark the performance of this GPU-accelerated eigensolver on two hybrid CPU–GPU architectures, namely a compute cluster based on Intel Xeon Gold CPUs and NVIDIA Volta GPUs, and the Summit supercomputer based on IBM POWER9 CPUs and NVIDIA Volta GPUs. Consistent with previous benchmarks on CPU-only architectures, the GPU-accelerated two-stage solver exhibits a parallel performance superior to the one-stage counterpart. Finally, we demonstrate the performance of the GPU-accelerated eigensolver developed in this work for routine semi-local KS-DFT calculations comprising thousands of atoms.

Název v anglickém jazyce

GPU-acceleration of the ELPA2 distributed eigensolver for dense symmetric and hermitian eigenproblems

Popis výsledku anglicky

The solution of eigenproblems is often a key computational bottleneck that limits the tractable system size of numerical algorithms, among them electronic structure theory in chemistry and in condensed matter physics. Large eigenproblems can easily exceed the capacity of a single compute node, thus must be solved on distributed-memory parallel computers. We here present GPU-oriented optimizations of the ELPA two-stage tridiagonalization eigensolver (ELPA2). On top of cuBLAS-based GPU offloading, we add a CUDA kernel to speed up the back-transformation of eigenvectors, which can be the computationally most expensive part of the two-stage tridiagonalization algorithm. We benchmark the performance of this GPU-accelerated eigensolver on two hybrid CPU–GPU architectures, namely a compute cluster based on Intel Xeon Gold CPUs and NVIDIA Volta GPUs, and the Summit supercomputer based on IBM POWER9 CPUs and NVIDIA Volta GPUs. Consistent with previous benchmarks on CPU-only architectures, the GPU-accelerated two-stage solver exhibits a parallel performance superior to the one-stage counterpart. Finally, we demonstrate the performance of the GPU-accelerated eigensolver developed in this work for routine semi-local KS-DFT calculations comprising thousands of atoms.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10101 - Pure mathematics

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2021

Kód důvěrnosti údajů

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

Název periodika

Computer Physics Communications

ISSN

0010-4655

e-ISSN

1879-2944

Svazek periodika

262

Číslo periodika v rámci svazku

May

Stát vydavatele periodika

NL - Nizozemsko

Počet stran výsledku

12

Strana od-do

107808

Kód UT WoS článku

000633365000004

EID výsledku v databázi Scopus

2-s2.0-85099623870