DC FieldValueLanguage
dc.contributor.authorArteaga, Andrea-
dc.contributor.authorRuprecht, Daniel-
dc.contributor.authorKrause, Rolf-
dc.date.accessioned2021-10-14T10:05:43Z-
dc.date.available2021-10-14T10:05:43Z-
dc.date.issued2015-09-15-
dc.identifier.citationApplied Mathematics and Computation 267: 727-741 (2015-09-15)de_DE
dc.identifier.issn0096-3003de_DE
dc.identifier.urihttp://hdl.handle.net/11420/10523-
dc.description.abstractIn view of the rapid rise of the number of cores in modern supercomputers, time-parallel methods that introduce concurrency along the temporal axis are becoming increasingly popular. For the solution of time-dependent partial differential equations, these methods can add another direction for concurrency on top of spatial parallelization. The paper presents an implementation of the time-parallel Parareal method in a C++ domain specific language for stencil computations (STELLA). STELLA provides both an OpenMP and a CUDA backend for a shared memory parallelization, using the CPU or GPU inside a node for the spatial stencils. Here, we intertwine this node-wise spatial parallelism with the time-parallel Parareal. This is done by adding an MPI-based implementation of Parareal, which allows us to parallelize in time across nodes. The performance of Parareal with both backends is analyzed in terms of speedup, parallel efficiency and energy-to-solution for an advection-diffusion problem with a time-dependent diffusion coefficient.en
dc.language.isoende_DE
dc.relation.ispartofApplied mathematics and computationde_DE
dc.subjectEnergy consumptionde_DE
dc.subjectParallel-in-timede_DE
dc.subjectPararealde_DE
dc.subjectSpeedupde_DE
dc.subjectSTELLAde_DE
dc.subjectStencil computationde_DE
dc.subjectComputer Science - Distributed; Parallel; and Cluster Computingde_DE
dc.subjectComputer Science - Distributed; Parallel; and Cluster Computingde_DE
dc.subjectMathematics - Numerical Analysisde_DE
dc.titleA stencil-based implementation of Parareal in the C++ domain specific embedded language STELLAde_DE
dc.typeArticlede_DE
dc.type.diniarticle-
dcterms.DCMITypeText-
tuhh.abstract.englishIn view of the rapid rise of the number of cores in modern supercomputers, time-parallel methods that introduce concurrency along the temporal axis are becoming increasingly popular. For the solution of time-dependent partial differential equations, these methods can add another direction for concurrency on top of spatial parallelization. The paper presents an implementation of the time-parallel Parareal method in a C++ domain specific language for stencil computations (STELLA). STELLA provides both an OpenMP and a CUDA backend for a shared memory parallelization, using the CPU or GPU inside a node for the spatial stencils. Here, we intertwine this node-wise spatial parallelism with the time-parallel Parareal. This is done by adding an MPI-based implementation of Parareal, which allows us to parallelize in time across nodes. The performance of Parareal with both backends is analyzed in terms of speedup, parallel efficiency and energy-to-solution for an advection-diffusion problem with a time-dependent diffusion coefficient.de_DE
tuhh.publisher.doi10.1016/j.amc.2014.12.055-
tuhh.type.opus(wissenschaftlicher) Artikel-
dc.type.driverarticle-
dc.type.casraiJournal Article-
tuhh.container.volume267de_DE
tuhh.container.startpage727de_DE
tuhh.container.endpage741de_DE
dc.identifier.arxiv1409.8563v2de_DE
dc.identifier.scopus2-s2.0-84942991344de_DE
local.publisher.peerreviewedtruede_DE
item.grantfulltextnone-
item.languageiso639-1en-
item.creatorOrcidArteaga, Andrea-
item.creatorOrcidRuprecht, Daniel-
item.creatorOrcidKrause, Rolf-
item.cerifentitytypePublications-
item.openairetypeArticle-
item.creatorGNDArteaga, Andrea-
item.creatorGNDRuprecht, Daniel-
item.creatorGNDKrause, Rolf-
item.mappedtypeArticle-
item.openairecristypehttp://purl.org/coar/resource_type/c_6501-
item.fulltextNo Fulltext-
crisitem.author.deptMathematik E-10-
crisitem.author.orcid0000-0003-1904-2473-
crisitem.author.parentorgStudiendekanat Elektrotechnik, Informatik und Mathematik-
Appears in Collections:Publications without fulltext
Show simple item record

Page view(s)

11
Last Week
0
Last month
checked on Nov 30, 2021

SCOPUSTM   
Citations

9
Last Week
0
Last month
checked on Nov 22, 2021

Google ScholarTM

Check

Add Files to Item

Note about this record

Cite this record

Export

Items in TORE are protected by copyright, with all rights reserved, unless otherwise indicated.