Acceleration Techniques for FETI Solvers for GPU Accelerators

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27740%2F18%3A10240250" target="_blank" >RIV/61989100:27740/18:10240250 - isvavai.cz</a>

Výsledek na webu

<a href="https://ieeexplore.ieee.org/document/8514396" target="_blank" >https://ieeexplore.ieee.org/document/8514396</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/HPCS.2018.00091" target="_blank" >10.1109/HPCS.2018.00091</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Acceleration Techniques for FETI Solvers for GPU Accelerators

Popis výsledku v původním jazyce

In this paper we evaluate several approaches to performing simultaneous matrix-vector multiplication of large numbers of matrices on a GPU accelerator. The goal of this evaluation is to develop efficient techniques for massively parallel Hybrid Total FETI solvers in our ESPRESO library. FETI solvers generally use sparse matrices. To overcome this we previously proposed the Local Schur Complement method for FETI to convert sparse matrices to their dense representation, without significantly increasing the memory requirements of the GPU accelerator. We selected the following techniques: standard GEMV, CUDA streams, dynamic parallelism, batched GEMM, BSR GEMV and HYB GEMV. Our results show that (i) if a FETI solver contains a large number of small matrices i.e. there is large number of small subdomains, then the best approach is dynamic parallelism; (ii) if there is small number of large subdomains, then the optimal approaches are dynamic parallelism and CUDA streams. Please note that Local Schur Complement method in conjunction with Hybrid Total FETI perform better with smaller subdomains.

Název v anglickém jazyce

Acceleration Techniques for FETI Solvers for GPU Accelerators

Popis výsledku anglicky

In this paper we evaluate several approaches to performing simultaneous matrix-vector multiplication of large numbers of matrices on a GPU accelerator. The goal of this evaluation is to develop efficient techniques for massively parallel Hybrid Total FETI solvers in our ESPRESO library. FETI solvers generally use sparse matrices. To overcome this we previously proposed the Local Schur Complement method for FETI to convert sparse matrices to their dense representation, without significantly increasing the memory requirements of the GPU accelerator. We selected the following techniques: standard GEMV, CUDA streams, dynamic parallelism, batched GEMM, BSR GEMV and HYB GEMV. Our results show that (i) if a FETI solver contains a large number of small matrices i.e. there is large number of small subdomains, then the best approach is dynamic parallelism; (ii) if there is small number of large subdomains, then the optimal approaches are dynamic parallelism and CUDA streams. Please note that Local Schur Complement method in conjunction with Hybrid Total FETI perform better with smaller subdomains.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

<a href="/cs/project/LM2015070" target="_blank" >LM2015070: IT4Innovations národní superpočítačové centrum</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2018

Kód důvěrnosti údajů

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

Název statě ve sborníku

2018 International Conference on High Performance Computing &amp; Simulation (HPCS) : proceedings

ISBN

978-1-5386-7879-4

ISSN

—

e-ISSN

neuvedeno

Počet stran výsledku

8

Strana od-do

546-553

Název nakladatele

IEEE

Místo vydání

Vienna

Místo konání akce

orelans

Datum konání akce

16. 7. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000450677700074