Multi-mass solvers for lattice QCD on GPUs

Graphical Processing Units (GPUs) are more and more frequently used for lattice QCD calculations. Lattice studies often require computing the quark propagators for several masses. These systems can be solved using multi-shift solvers but these algorithms are memory intensive which limits the size of...

Full description

Saved in:
Bibliographic Details
Published in:Journal of computational physics 2012-02, Vol.231 (4), p.1866-1878
Main Authors: Alexandru, A., Pelissier, C., Gamari, B., Lee, F.X.
Format: Article
Language:English
Subjects:
Algorithms
BiCGstab
BiCGstab-M
Computation
Computational techniques
Exact sciences and technology
Fermions
GPU
Lattices
Mathematical analysis
Mathematical methods in physics
Physics
Quarks
Solvers
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c360t-d1ceedb827fcc77cedf07bbd326f4fe008a5c04811b8131983b10c8a287523c23
cites cdi_FETCH-LOGICAL-c360t-d1ceedb827fcc77cedf07bbd326f4fe008a5c04811b8131983b10c8a287523c23
container_end_page 1878
container_issue 4
container_start_page 1866
container_title Journal of computational physics
container_volume 231
creator Alexandru, A.
Pelissier, C.
Gamari, B.
Lee, F.X.
description Graphical Processing Units (GPUs) are more and more frequently used for lattice QCD calculations. Lattice studies often require computing the quark propagators for several masses. These systems can be solved using multi-shift solvers but these algorithms are memory intensive which limits the size of the problem that can be solved using GPUs. In this paper, we show how to efficiently use a memory-lean single-mass solver to solve multi-mass problems. We focus on the BiCGstab algorithm for Wilson fermions and show that the single-mass solver not only requires less memory but also outperforms the multi-shift variant by a factor of two.
doi_str_mv 10.1016/j.jcp.2011.11.003
format article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_1010890068</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0021999111006450</els_id><sourcerecordid>1010890068</sourcerecordid><originalsourceid>FETCH-LOGICAL-c360t-d1ceedb827fcc77cedf07bbd326f4fe008a5c04811b8131983b10c8a287523c23</originalsourceid><addsrcrecordid>eNp9kFFLwzAUhYMoOKc_wLe-CCK03qRrm-CTTJ3CRAX3HNLbBFK6duZ2A_-9GRMfhQP35Tvncg5jlxwyDry8bbMWN5kAzrMogPyITTgoSEXFy2M2ARA8VUrxU3ZG1AKALGZywm5et93o07UhSmjodjZQ4oaQdGYcPdrkY_6QDH2yeF_ROTtxpiN78XunbPX0-Dl_Tpdvi5f5_TLFvIQxbTha29RSVA6xqtA2Dqq6bnJRupmz8bEpEGaS81rynCuZ1xxQGiGrQuQo8im7PuRuwvC1tTTqtSe0XWd6O2xJx74gFUApI8oPKIaBKFinN8GvTfiO0J4rdavjLnq_i46Ku0TP1W-8ITSdC6ZHT39GUSihClFF7u7A2dh1523QhN72sY8PFkfdDP6fLz-jJHWd</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1010890068</pqid></control><display><type>article</type><title>Multi-mass solvers for lattice QCD on GPUs</title><source>Elsevier</source><creator>Alexandru, A. ; Pelissier, C. ; Gamari, B. ; Lee, F.X.</creator><creatorcontrib>Alexandru, A. ; Pelissier, C. ; Gamari, B. ; Lee, F.X.</creatorcontrib><description>Graphical Processing Units (GPUs) are more and more frequently used for lattice QCD calculations. Lattice studies often require computing the quark propagators for several masses. These systems can be solved using multi-shift solvers but these algorithms are memory intensive which limits the size of the problem that can be solved using GPUs. In this paper, we show how to efficiently use a memory-lean single-mass solver to solve multi-mass problems. We focus on the BiCGstab algorithm for Wilson fermions and show that the single-mass solver not only requires less memory but also outperforms the multi-shift variant by a factor of two.</description><identifier>ISSN: 0021-9991</identifier><identifier>EISSN: 1090-2716</identifier><identifier>DOI: 10.1016/j.jcp.2011.11.003</identifier><identifier>CODEN: JCTPAH</identifier><language>eng</language><publisher>Kidlington: Elsevier Inc</publisher><subject>Algorithms ; BiCGstab ; BiCGstab-M ; Computation ; Computational techniques ; Exact sciences and technology ; Fermions ; GPU ; Lattices ; Mathematical analysis ; Mathematical methods in physics ; Physics ; Quarks ; Solvers</subject><ispartof>Journal of computational physics, 2012-02, Vol.231 (4), p.1866-1878</ispartof><rights>2011 Elsevier Inc.</rights><rights>2015 INIST-CNRS</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c360t-d1ceedb827fcc77cedf07bbd326f4fe008a5c04811b8131983b10c8a287523c23</citedby><cites>FETCH-LOGICAL-c360t-d1ceedb827fcc77cedf07bbd326f4fe008a5c04811b8131983b10c8a287523c23</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&amp;idt=25929527$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><creatorcontrib>Alexandru, A.</creatorcontrib><creatorcontrib>Pelissier, C.</creatorcontrib><creatorcontrib>Gamari, B.</creatorcontrib><creatorcontrib>Lee, F.X.</creatorcontrib><title>Multi-mass solvers for lattice QCD on GPUs</title><title>Journal of computational physics</title><description>Graphical Processing Units (GPUs) are more and more frequently used for lattice QCD calculations. Lattice studies often require computing the quark propagators for several masses. These systems can be solved using multi-shift solvers but these algorithms are memory intensive which limits the size of the problem that can be solved using GPUs. In this paper, we show how to efficiently use a memory-lean single-mass solver to solve multi-mass problems. We focus on the BiCGstab algorithm for Wilson fermions and show that the single-mass solver not only requires less memory but also outperforms the multi-shift variant by a factor of two.</description><subject>Algorithms</subject><subject>BiCGstab</subject><subject>BiCGstab-M</subject><subject>Computation</subject><subject>Computational techniques</subject><subject>Exact sciences and technology</subject><subject>Fermions</subject><subject>GPU</subject><subject>Lattices</subject><subject>Mathematical analysis</subject><subject>Mathematical methods in physics</subject><subject>Physics</subject><subject>Quarks</subject><subject>Solvers</subject><issn>0021-9991</issn><issn>1090-2716</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2012</creationdate><recordtype>article</recordtype><recordid>eNp9kFFLwzAUhYMoOKc_wLe-CCK03qRrm-CTTJ3CRAX3HNLbBFK6duZ2A_-9GRMfhQP35Tvncg5jlxwyDry8bbMWN5kAzrMogPyITTgoSEXFy2M2ARA8VUrxU3ZG1AKALGZywm5et93o07UhSmjodjZQ4oaQdGYcPdrkY_6QDH2yeF_ROTtxpiN78XunbPX0-Dl_Tpdvi5f5_TLFvIQxbTha29RSVA6xqtA2Dqq6bnJRupmz8bEpEGaS81rynCuZ1xxQGiGrQuQo8im7PuRuwvC1tTTqtSe0XWd6O2xJx74gFUApI8oPKIaBKFinN8GvTfiO0J4rdavjLnq_i46Ku0TP1W-8ITSdC6ZHT39GUSihClFF7u7A2dh1523QhN72sY8PFkfdDP6fLz-jJHWd</recordid><startdate>20120220</startdate><enddate>20120220</enddate><creator>Alexandru, A.</creator><creator>Pelissier, C.</creator><creator>Gamari, B.</creator><creator>Lee, F.X.</creator><general>Elsevier Inc</general><general>Elsevier</general><scope>IQODW</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>7U5</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>20120220</creationdate><title>Multi-mass solvers for lattice QCD on GPUs</title><author>Alexandru, A. ; Pelissier, C. ; Gamari, B. ; Lee, F.X.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c360t-d1ceedb827fcc77cedf07bbd326f4fe008a5c04811b8131983b10c8a287523c23</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Algorithms</topic><topic>BiCGstab</topic><topic>BiCGstab-M</topic><topic>Computation</topic><topic>Computational techniques</topic><topic>Exact sciences and technology</topic><topic>Fermions</topic><topic>GPU</topic><topic>Lattices</topic><topic>Mathematical analysis</topic><topic>Mathematical methods in physics</topic><topic>Physics</topic><topic>Quarks</topic><topic>Solvers</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Alexandru, A.</creatorcontrib><creatorcontrib>Pelissier, C.</creatorcontrib><creatorcontrib>Gamari, B.</creatorcontrib><creatorcontrib>Lee, F.X.</creatorcontrib><collection>Pascal-Francis</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics &amp; Communications Abstracts</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><jtitle>Journal of computational physics</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Alexandru, A.</au><au>Pelissier, C.</au><au>Gamari, B.</au><au>Lee, F.X.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Multi-mass solvers for lattice QCD on GPUs</atitle><jtitle>Journal of computational physics</jtitle><date>2012-02-20</date><risdate>2012</risdate><volume>231</volume><issue>4</issue><spage>1866</spage><epage>1878</epage><pages>1866-1878</pages><issn>0021-9991</issn><eissn>1090-2716</eissn><coden>JCTPAH</coden><abstract>Graphical Processing Units (GPUs) are more and more frequently used for lattice QCD calculations. Lattice studies often require computing the quark propagators for several masses. These systems can be solved using multi-shift solvers but these algorithms are memory intensive which limits the size of the problem that can be solved using GPUs. In this paper, we show how to efficiently use a memory-lean single-mass solver to solve multi-mass problems. We focus on the BiCGstab algorithm for Wilson fermions and show that the single-mass solver not only requires less memory but also outperforms the multi-shift variant by a factor of two.</abstract><cop>Kidlington</cop><pub>Elsevier Inc</pub><doi>10.1016/j.jcp.2011.11.003</doi><tpages>13</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0021-9991
ispartof Journal of computational physics, 2012-02, Vol.231 (4), p.1866-1878
issn 0021-9991
1090-2716
language eng
recordid cdi_proquest_miscellaneous_1010890068
source Elsevier
subjects Algorithms
BiCGstab
BiCGstab-M
Computation
Computational techniques
Exact sciences and technology
Fermions
GPU
Lattices
Mathematical analysis
Mathematical methods in physics
Physics
Quarks
Solvers
title Multi-mass solvers for lattice QCD on GPUs
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-27T15%3A44%3A23IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Multi-mass%20solvers%20for%20lattice%20QCD%20on%20GPUs&rft.jtitle=Journal%20of%20computational%20physics&rft.au=Alexandru,%20A.&rft.date=2012-02-20&rft.volume=231&rft.issue=4&rft.spage=1866&rft.epage=1878&rft.pages=1866-1878&rft.issn=0021-9991&rft.eissn=1090-2716&rft.coden=JCTPAH&rft_id=info:doi/10.1016/j.jcp.2011.11.003&rft_dat=%3Cproquest_cross%3E1010890068%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c360t-d1ceedb827fcc77cedf07bbd326f4fe008a5c04811b8131983b10c8a287523c23%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=1010890068&rft_id=info:pmid/&rfr_iscdi=true