Benchmarking GPU and CPU codes for Heisenberg spin glass over-relaxation

We present a set of possible implementations for Graphics Processing Units (GPU) of the Over-relaxation technique applied to the 3D Heisenberg spin glass model. The results show that a carefully tuned code can achieve more than 100 GFlops/s of sustained performance and update a single spin in about...

Full description

Saved in:
Bibliographic Details
Published in:Computer physics communications 2011-06, Vol.182 (6), p.1265-1271
Main Authors: Bernaschi, M., Parisi, G., Parisi, L.
Format: Article
Language:English
Subjects:
Benchmarking
Central processing units
Computer simulation
GPU
Mathematical models
Nanocomposites
Nanomaterials
Nanostructure
Spin glass
Spin systems
Three dimensional
Vector processing
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-c396t-a19959d10e7ada34dc2ea3001ae9b1c5b28cf9858b89fbd97131b99b23243b073
cites cdi_FETCH-LOGICAL-c396t-a19959d10e7ada34dc2ea3001ae9b1c5b28cf9858b89fbd97131b99b23243b073
container_end_page 1271
container_issue 6
container_start_page 1265
container_title Computer physics communications
container_volume 182
creator Bernaschi, M.
Parisi, G.
Parisi, L.
description We present a set of possible implementations for Graphics Processing Units (GPU) of the Over-relaxation technique applied to the 3D Heisenberg spin glass model. The results show that a carefully tuned code can achieve more than 100 GFlops/s of sustained performance and update a single spin in about 0.6 nanoseconds. A multi-hit technique that exploits the GPU shared memory further reduces this time. Such results are compared with those obtained by means of a highly-tuned vector-parallel code on latest generation multi-core CPUs.
doi_str_mv 10.1016/j.cpc.2011.02.012
format article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_1671246621</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0010465511000749</els_id><sourcerecordid>1671246621</sourcerecordid><originalsourceid>FETCH-LOGICAL-c396t-a19959d10e7ada34dc2ea3001ae9b1c5b28cf9858b89fbd97131b99b23243b073</originalsourceid><addsrcrecordid>eNp9kD1PwzAURS0EEqXwA9g8siT42fmymKCCFqkSDHS2HPuluKRxsVME_x5XZWa6yz1P7x5CroHlwKC63eRmZ3LOAHLGcwb8hEygqWXGZVGckgljwLKiKstzchHjhjFW11JMyOIBB_O-1eHDDWs6f11RPVg6S2m8xUg7H-gCXcShxbCmcecGuu51jNR_YcgC9vpbj84Pl-Ss033Eq7-cktXT49tskS1f5s-z-2VmhKzGTIOUpbTAsNZWi8Iajlqk7zTKFkzZ8sZ0simbtpFda2UNAlopWy54IVpWiym5Od7dBf-5xziqrYsG-14P6PdRQVUDL6qKQ6rCsWqCjzFgp3bBpak_Cpg6WFMblaypgzXFuErWEnN3ZDBt-HIYVDQuKULrAppRWe_-oX8BT1dz4g</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1671246621</pqid></control><display><type>article</type><title>Benchmarking GPU and CPU codes for Heisenberg spin glass over-relaxation</title><source>ScienceDirect Freedom Collection 2022-2024</source><creator>Bernaschi, M. ; Parisi, G. ; Parisi, L.</creator><creatorcontrib>Bernaschi, M. ; Parisi, G. ; Parisi, L.</creatorcontrib><description>We present a set of possible implementations for Graphics Processing Units (GPU) of the Over-relaxation technique applied to the 3D Heisenberg spin glass model. The results show that a carefully tuned code can achieve more than 100 GFlops/s of sustained performance and update a single spin in about 0.6 nanoseconds. A multi-hit technique that exploits the GPU shared memory further reduces this time. Such results are compared with those obtained by means of a highly-tuned vector-parallel code on latest generation multi-core CPUs.</description><identifier>ISSN: 0010-4655</identifier><identifier>EISSN: 1879-2944</identifier><identifier>EISSN: 1386-9485</identifier><identifier>DOI: 10.1016/j.cpc.2011.02.012</identifier><language>eng</language><publisher>Elsevier B.V</publisher><subject>Benchmarking ; Central processing units ; Computer simulation ; GPU ; Mathematical models ; Nanocomposites ; Nanomaterials ; Nanostructure ; Spin glass ; Spin systems ; Three dimensional ; Vector processing</subject><ispartof>Computer physics communications, 2011-06, Vol.182 (6), p.1265-1271</ispartof><rights>2011 Elsevier B.V.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c396t-a19959d10e7ada34dc2ea3001ae9b1c5b28cf9858b89fbd97131b99b23243b073</citedby><cites>FETCH-LOGICAL-c396t-a19959d10e7ada34dc2ea3001ae9b1c5b28cf9858b89fbd97131b99b23243b073</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></links><search><creatorcontrib>Bernaschi, M.</creatorcontrib><creatorcontrib>Parisi, G.</creatorcontrib><creatorcontrib>Parisi, L.</creatorcontrib><title>Benchmarking GPU and CPU codes for Heisenberg spin glass over-relaxation</title><title>Computer physics communications</title><description>We present a set of possible implementations for Graphics Processing Units (GPU) of the Over-relaxation technique applied to the 3D Heisenberg spin glass model. The results show that a carefully tuned code can achieve more than 100 GFlops/s of sustained performance and update a single spin in about 0.6 nanoseconds. A multi-hit technique that exploits the GPU shared memory further reduces this time. Such results are compared with those obtained by means of a highly-tuned vector-parallel code on latest generation multi-core CPUs.</description><subject>Benchmarking</subject><subject>Central processing units</subject><subject>Computer simulation</subject><subject>GPU</subject><subject>Mathematical models</subject><subject>Nanocomposites</subject><subject>Nanomaterials</subject><subject>Nanostructure</subject><subject>Spin glass</subject><subject>Spin systems</subject><subject>Three dimensional</subject><subject>Vector processing</subject><issn>0010-4655</issn><issn>1879-2944</issn><issn>1386-9485</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2011</creationdate><recordtype>article</recordtype><recordid>eNp9kD1PwzAURS0EEqXwA9g8siT42fmymKCCFqkSDHS2HPuluKRxsVME_x5XZWa6yz1P7x5CroHlwKC63eRmZ3LOAHLGcwb8hEygqWXGZVGckgljwLKiKstzchHjhjFW11JMyOIBB_O-1eHDDWs6f11RPVg6S2m8xUg7H-gCXcShxbCmcecGuu51jNR_YcgC9vpbj84Pl-Ss033Eq7-cktXT49tskS1f5s-z-2VmhKzGTIOUpbTAsNZWi8Iajlqk7zTKFkzZ8sZ0simbtpFda2UNAlopWy54IVpWiym5Od7dBf-5xziqrYsG-14P6PdRQVUDL6qKQ6rCsWqCjzFgp3bBpak_Cpg6WFMblaypgzXFuErWEnN3ZDBt-HIYVDQuKULrAppRWe_-oX8BT1dz4g</recordid><startdate>20110601</startdate><enddate>20110601</enddate><creator>Bernaschi, M.</creator><creator>Parisi, G.</creator><creator>Parisi, L.</creator><general>Elsevier B.V</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7U5</scope><scope>8FD</scope><scope>H8D</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>20110601</creationdate><title>Benchmarking GPU and CPU codes for Heisenberg spin glass over-relaxation</title><author>Bernaschi, M. ; Parisi, G. ; Parisi, L.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c396t-a19959d10e7ada34dc2ea3001ae9b1c5b28cf9858b89fbd97131b99b23243b073</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2011</creationdate><topic>Benchmarking</topic><topic>Central processing units</topic><topic>Computer simulation</topic><topic>GPU</topic><topic>Mathematical models</topic><topic>Nanocomposites</topic><topic>Nanomaterials</topic><topic>Nanostructure</topic><topic>Spin glass</topic><topic>Spin systems</topic><topic>Three dimensional</topic><topic>Vector processing</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Bernaschi, M.</creatorcontrib><creatorcontrib>Parisi, G.</creatorcontrib><creatorcontrib>Parisi, L.</creatorcontrib><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>Technology Research Database</collection><collection>Aerospace 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>Computer physics communications</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Bernaschi, M.</au><au>Parisi, G.</au><au>Parisi, L.</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Benchmarking GPU and CPU codes for Heisenberg spin glass over-relaxation</atitle><jtitle>Computer physics communications</jtitle><date>2011-06-01</date><risdate>2011</risdate><volume>182</volume><issue>6</issue><spage>1265</spage><epage>1271</epage><pages>1265-1271</pages><issn>0010-4655</issn><eissn>1879-2944</eissn><eissn>1386-9485</eissn><abstract>We present a set of possible implementations for Graphics Processing Units (GPU) of the Over-relaxation technique applied to the 3D Heisenberg spin glass model. The results show that a carefully tuned code can achieve more than 100 GFlops/s of sustained performance and update a single spin in about 0.6 nanoseconds. A multi-hit technique that exploits the GPU shared memory further reduces this time. Such results are compared with those obtained by means of a highly-tuned vector-parallel code on latest generation multi-core CPUs.</abstract><pub>Elsevier B.V</pub><doi>10.1016/j.cpc.2011.02.012</doi><tpages>7</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0010-4655
ispartof Computer physics communications, 2011-06, Vol.182 (6), p.1265-1271
issn 0010-4655
1879-2944
1386-9485
language eng
recordid cdi_proquest_miscellaneous_1671246621
source ScienceDirect Freedom Collection 2022-2024
subjects Benchmarking
Central processing units
Computer simulation
GPU
Mathematical models
Nanocomposites
Nanomaterials
Nanostructure
Spin glass
Spin systems
Three dimensional
Vector processing
title Benchmarking GPU and CPU codes for Heisenberg spin glass over-relaxation
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-07T07%3A31%3A57IST&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=Benchmarking%20GPU%20and%20CPU%20codes%20for%20Heisenberg%20spin%20glass%20over-relaxation&rft.jtitle=Computer%20physics%20communications&rft.au=Bernaschi,%20M.&rft.date=2011-06-01&rft.volume=182&rft.issue=6&rft.spage=1265&rft.epage=1271&rft.pages=1265-1271&rft.issn=0010-4655&rft.eissn=1879-2944&rft_id=info:doi/10.1016/j.cpc.2011.02.012&rft_dat=%3Cproquest_cross%3E1671246621%3C/proquest_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c396t-a19959d10e7ada34dc2ea3001ae9b1c5b28cf9858b89fbd97131b99b23243b073%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=1671246621&rft_id=info:pmid/&rfr_iscdi=true