Acceleration of tensor-product operations for high-order finite element methods

Simple item page
dc.contributor.authorƚwirydowicz, K.en
dc.contributor.authorChalmers, N.en
dc.contributor.authorKarakus, A.en
dc.contributor.authorWarburton, T.en
dc.contributor.departmentMathematicsen
dc.date.accessioned2018-01-19T12:55:12Zen
dc.date.available2018-01-19T12:55:12Zen
dc.date.issued2017-09en
dc.description.abstractThis paper is devoted to GPU kernel optimization and performance analysis of three tensor-product operators arising in finite element methods. We provide a mathematical background to these operations and implementation details. Achieving close-to-the-peak performance for these operators requires extensive optimization because of the operators' properties: low arithmetic intensity, tiered structure, and the need to store intermediate results inside the kernel. We give a guided overview of optimization strategies and we present a performance model that allows us to compare the efficacy of these optimizations against an empirically calibrated roofline.en
dc.description.notes31 pages, 11 figuresen
dc.identifier.orcidWarburton, T [0000-0002-3202-1151]en
dc.identifier.urihttp://hdl.handle.net/10919/81866en
dc.language.isoenen
dc.relation.urihttp://arxiv.org/abs/1711.00903v2en
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectcs.MSen
dc.subjectcs.DCen
dc.subjectcs.NAen
dc.subjectcs.PFen
dc.subjectmath.NAen
dc.titleAcceleration of tensor-product operations for high-order finite element methodsen
dc.typeArticle - Refereeden
pubs.organisational-group/Virginia Techen
pubs.organisational-group/Virginia Tech/All T&R Facultyen
pubs.organisational-group/Virginia Tech/Scienceen
pubs.organisational-group/Virginia Tech/Science/COS T&R Facultyen
pubs.organisational-group/Virginia Tech/Science/Mathematicsen

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
1711.00903v2.pdf
Size:
6.01 MB
Format:
Adobe Portable Document Format
Description:
Draft Version
Download
License bundle
Now showing 1 - 1 of 1
Name:
VTUL_Distribution_License_2016_05_09.pdf
Size:
18.09 KB
Format:
Adobe Portable Document Format
Description:
Download

Collections

All Faculty Deposits
Scholarly Works, Mathematics