The performances of R GPU implementations of the GMRES method
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989592%3A15210%2F18%3A73587584" target="_blank" >RIV/61989592:15210/18:73587584 - isvavai.cz</a>
Result on the web
<a href="http://www.revistadestatistica.ro/wp-content/uploads/2018/03/RRS_1_2018_A09.pdf" target="_blank" >http://www.revistadestatistica.ro/wp-content/uploads/2018/03/RRS_1_2018_A09.pdf</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
The performances of R GPU implementations of the GMRES method
Original language description
Although the performance of commodity computers has improved drastically with the introduction of multicore processors and GPU computing, the standard R distribution is still based on single-threaded model of computation, using only a small fraction of the computational power available now for most desktops and laptops. Modern statistical software packages rely on high performance implementations of the linear algebra routines there are at the core of several important leading edge statistical methods. In this paper we present a GPU implementation of the GMRES iterative method for solving linear systems. We compare the performance of this implementation with a pure single threaded version of the CPU. We also investigate the performance of our implementation using different GPU packages available now for R such as gmatrix, gputools or gpuR which are based on CUDA or OpenCL frameworks.
Czech name
—
Czech description
—
Classification
Type
J<sub>imp</sub> - Article in a specialist periodical, which is included in the Web of Science database
CEP classification
—
OECD FORD branch
50202 - Applied Economics, Econometrics
Result continuities
Project
—
Continuities
I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace
Others
Publication year
2018
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Name of the periodical
Romanian Statistical Review
ISSN
1018-046X
e-ISSN
—
Volume of the periodical
2018
Issue of the periodical within the volume
1
Country of publishing house
RO - ROMANIA
Number of pages
12
Pages from-to
121-132
UT code for WoS article
000429314700009
EID of the result in the Scopus database
—