Fast Multipole Method as a Matrix-Free Hierarchical Low-Rank Approximation

Handle URI:
http://hdl.handle.net/10754/627144
Title:
Fast Multipole Method as a Matrix-Free Hierarchical Low-Rank Approximation
Authors:
Yokota, Rio; Ibeid, Huda ( 0000-0001-5208-5366 ) ; Keyes, David E. ( 0000-0002-4052-7224 )
Abstract:
There has been a large increase in the amount of work on hierarchical low-rank approximation methods, where the interest is shared by multiple communities that previously did not intersect. This objective of this article is two-fold; to provide a thorough review of the recent advancements in this field from both analytical and algebraic perspectives, and to present a comparative benchmark of two highly optimized implementations of contrasting methods for some simple yet representative test cases. The first half of this paper has the form of a survey paper, to achieve the former objective. We categorize the recent advances in this field from the perspective of compute-memory tradeoff, which has not been considered in much detail in this area. Benchmark tests reveal that there is a large difference in the memory consumption and performance between the different methods.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Computer Science Program; Applied Mathematics and Computational Science Program; Extreme Computing Research Center
Citation:
Yokota R, Ibeid H, Keyes D (2017) Fast Multipole Method as a Matrix-Free Hierarchical Low-Rank Approximation. Eigenvalue Problems: Algorithms, Software and Applications in Petascale Computing: 267–286. Available: http://dx.doi.org/10.1007/978-3-319-62426-6_17.
Publisher:
Springer International Publishing
Journal:
Eigenvalue Problems: Algorithms, Software and Applications in Petascale Computing
KAUST Grant Number:
KUK-C1-013-04
Conference/Event name:
1st InternationalWorkshop on Eigenvalue Problems: Algorithms, Software and Applications in Petascale Computing, EPASA 2015
Issue Date:
3-Jan-2018
DOI:
10.1007/978-3-319-62426-6_17
Type:
Conference Paper
ISSN:
1439-7358; 2197-7100
Sponsors:
We thank François-Henry Rouet, Pieter Ghysels, and Xiaoye, S. Li for providing the STRUMPACK interface for our comparisons between FMM and HSS. This work was supported by JSPS KAKENHI Grant-in-Aid for Research Activity Start-up Grant Number 15H06196. This publication was based on work supported in part by Award No KUK-C1-013-04, made by King Abdullah University of Science and Technology (KAUST). This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number OCI-1053575.
Additional Links:
https://link.springer.com/chapter/10.1007%2F978-3-319-62426-6_17
Appears in Collections:
Conference Papers; Applied Mathematics and Computational Science Program; Extreme Computing Research Center; Computer Science Program; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorYokota, Rioen
dc.contributor.authorIbeid, Hudaen
dc.contributor.authorKeyes, David E.en
dc.date.accessioned2018-02-19T09:56:36Z-
dc.date.available2018-02-19T09:56:36Z-
dc.date.issued2018-01-03en
dc.identifier.citationYokota R, Ibeid H, Keyes D (2017) Fast Multipole Method as a Matrix-Free Hierarchical Low-Rank Approximation. Eigenvalue Problems: Algorithms, Software and Applications in Petascale Computing: 267–286. Available: http://dx.doi.org/10.1007/978-3-319-62426-6_17.en
dc.identifier.issn1439-7358en
dc.identifier.issn2197-7100en
dc.identifier.doi10.1007/978-3-319-62426-6_17en
dc.identifier.urihttp://hdl.handle.net/10754/627144-
dc.description.abstractThere has been a large increase in the amount of work on hierarchical low-rank approximation methods, where the interest is shared by multiple communities that previously did not intersect. This objective of this article is two-fold; to provide a thorough review of the recent advancements in this field from both analytical and algebraic perspectives, and to present a comparative benchmark of two highly optimized implementations of contrasting methods for some simple yet representative test cases. The first half of this paper has the form of a survey paper, to achieve the former objective. We categorize the recent advances in this field from the perspective of compute-memory tradeoff, which has not been considered in much detail in this area. Benchmark tests reveal that there is a large difference in the memory consumption and performance between the different methods.en
dc.description.sponsorshipWe thank François-Henry Rouet, Pieter Ghysels, and Xiaoye, S. Li for providing the STRUMPACK interface for our comparisons between FMM and HSS. This work was supported by JSPS KAKENHI Grant-in-Aid for Research Activity Start-up Grant Number 15H06196. This publication was based on work supported in part by Award No KUK-C1-013-04, made by King Abdullah University of Science and Technology (KAUST). This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number OCI-1053575.en
dc.publisherSpringer International Publishingen
dc.relation.urlhttps://link.springer.com/chapter/10.1007%2F978-3-319-62426-6_17en
dc.rightsThe final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-62426-6_17en
dc.titleFast Multipole Method as a Matrix-Free Hierarchical Low-Rank Approximationen
dc.typeConference Paperen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentComputer Science Programen
dc.contributor.departmentApplied Mathematics and Computational Science Programen
dc.contributor.departmentExtreme Computing Research Centeren
dc.identifier.journalEigenvalue Problems: Algorithms, Software and Applications in Petascale Computingen
dc.conference.date2015-09-14 to 2015-09-16en
dc.conference.name1st InternationalWorkshop on Eigenvalue Problems: Algorithms, Software and Applications in Petascale Computing, EPASA 2015en
dc.conference.locationTsukuba, JPNen
dc.eprint.versionPost-printen
dc.contributor.institutionTokyo Institute of Technology, 2-12-1 O-okayama Meguro-ku, Tokyo, , Japanen
kaust.authorIbeid, Hudaen
kaust.authorKeyes, David E.en
kaust.grant.numberKUK-C1-013-04en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.