Performance of SOR methods on modern vector and scalar processors
The building-cube method (BCM) is a new generation algorithm for CFD simulations. The basic idea of BCM is to simplify the algorithm in all stages of flow computation to achieve large-scale simulations. Calculation of a pressure field using the Successive Over Relaxation (SOR) method consumes most o...
Gespeichert in:
| Veröffentlicht in: | Computers & fluids 2011-06, Vol.45 (1), p.215-221 |
|---|---|
| Hauptverfasser: | , , , , , , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 221 |
|---|---|
| container_issue | 1 |
| container_start_page | 215 |
| container_title | Computers & fluids |
| container_volume | 45 |
| creator | Soga, Takashi Musa, Akihiro Okabe, Koki Komatsu, Kazuhiko Egawa, Ryusuke Takizawa, Hiroyuki Kobayashi, Hiroaki Takahashi, Shun Sasaki, Daisuke Nakahashi, Kazuhiro |
| description | The building-cube method (BCM) is a new generation algorithm for CFD simulations. The basic idea of BCM is to simplify the algorithm in all stages of flow computation to achieve large-scale simulations. Calculation of a pressure field using the Successive Over Relaxation (SOR) method consumes most of the total execution time required for BCM. In this paper, effective implementations on modern vector and scalar processors are investigated. NEC SX-9 and Intel Nehalem-EX are the latest vector and scalar processors. Those processors have much higher peak performances than their previous-generation processors. However, their memory bandwidth improvement cannot catch up with the performance improvement of processors. This is the so-called memory wall problem. In our paper, we discuss optimization techniques for implementation of the SOR method based on architectural characteristics of these modern processors, and evaluate their effects on the sustained performances of these processors for BCM. |
| doi_str_mv | 10.1016/j.compfluid.2010.12.024 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_875032689</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0045793010003804</els_id><sourcerecordid>875032689</sourcerecordid><originalsourceid>FETCH-LOGICAL-c446t-9e60a660a84acf7b91cf7a5400841a1335a2aee439b2e71c21b731bffc2643d03</originalsourceid><addsrcrecordid>eNqFkE1PwzAMhiMEEmPwG8gNLi3OR5v2OE18SZOG-DhHaeqITm0zkg6Jf0-mIY5wsC1br1_LDyGXDHIGrLzZ5NYPW9fvujbnsJ_yHLg8IjNWqToDJdUxmQHIIlO1gFNyFuMGUi-4nJHFEwbnw2BGi9Q7-rJ-pgNO776N1I908C2GkX6inXygZmxptKY3gW6DtxijD_GcnDjTR7z4qXPydnf7unzIVuv7x-VilVkpyymrsQRTpqiksU41NUvZFBKgkswwIQrDDaIUdcNRMctZowRrnLO8lKIFMSdXB990-mOHcdJDFy32vRnR76KuVAGCl1WdlNd_KpmCWlWlUEWSqoPUBh9jQKe3oRtM-NIM9B6v3uhfvHqPVzOuE960uThsYvr5s8Ogo-0wUWy7kGjp1nf_enwDHK6HIw</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1709786375</pqid></control><display><type>article</type><title>Performance of SOR methods on modern vector and scalar processors</title><source>Elsevier ScienceDirect Journals Complete</source><creator>Soga, Takashi ; Musa, Akihiro ; Okabe, Koki ; Komatsu, Kazuhiko ; Egawa, Ryusuke ; Takizawa, Hiroyuki ; Kobayashi, Hiroaki ; Takahashi, Shun ; Sasaki, Daisuke ; Nakahashi, Kazuhiro</creator><creatorcontrib>Soga, Takashi ; Musa, Akihiro ; Okabe, Koki ; Komatsu, Kazuhiko ; Egawa, Ryusuke ; Takizawa, Hiroyuki ; Kobayashi, Hiroaki ; Takahashi, Shun ; Sasaki, Daisuke ; Nakahashi, Kazuhiro</creatorcontrib><description>The building-cube method (BCM) is a new generation algorithm for CFD simulations. The basic idea of BCM is to simplify the algorithm in all stages of flow computation to achieve large-scale simulations. Calculation of a pressure field using the Successive Over Relaxation (SOR) method consumes most of the total execution time required for BCM. In this paper, effective implementations on modern vector and scalar processors are investigated. NEC SX-9 and Intel Nehalem-EX are the latest vector and scalar processors. Those processors have much higher peak performances than their previous-generation processors. However, their memory bandwidth improvement cannot catch up with the performance improvement of processors. This is the so-called memory wall problem. In our paper, we discuss optimization techniques for implementation of the SOR method based on architectural characteristics of these modern processors, and evaluate their effects on the sustained performances of these processors for BCM.</description><identifier>ISSN: 0045-7930</identifier><identifier>EISSN: 1879-0747</identifier><identifier>DOI: 10.1016/j.compfluid.2010.12.024</identifier><language>eng</language><publisher>Elsevier Ltd</publisher><subject>Algorithms ; Building-cube method ; Computational fluid dynamics ; Computer simulation ; Mathematical analysis ; Mathematical models ; Processors ; Scalars ; SOR method ; Vector and scalar processing ; Vectors (mathematics)</subject><ispartof>Computers & fluids, 2011-06, Vol.45 (1), p.215-221</ispartof><rights>2010 Elsevier Ltd</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c446t-9e60a660a84acf7b91cf7a5400841a1335a2aee439b2e71c21b731bffc2643d03</citedby><cites>FETCH-LOGICAL-c446t-9e60a660a84acf7b91cf7a5400841a1335a2aee439b2e71c21b731bffc2643d03</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.sciencedirect.com/science/article/pii/S0045793010003804$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>314,776,780,3537,27901,27902,65306</link.rule.ids></links><search><creatorcontrib>Soga, Takashi</creatorcontrib><creatorcontrib>Musa, Akihiro</creatorcontrib><creatorcontrib>Okabe, Koki</creatorcontrib><creatorcontrib>Komatsu, Kazuhiko</creatorcontrib><creatorcontrib>Egawa, Ryusuke</creatorcontrib><creatorcontrib>Takizawa, Hiroyuki</creatorcontrib><creatorcontrib>Kobayashi, Hiroaki</creatorcontrib><creatorcontrib>Takahashi, Shun</creatorcontrib><creatorcontrib>Sasaki, Daisuke</creatorcontrib><creatorcontrib>Nakahashi, Kazuhiro</creatorcontrib><title>Performance of SOR methods on modern vector and scalar processors</title><title>Computers & fluids</title><description>The building-cube method (BCM) is a new generation algorithm for CFD simulations. The basic idea of BCM is to simplify the algorithm in all stages of flow computation to achieve large-scale simulations. Calculation of a pressure field using the Successive Over Relaxation (SOR) method consumes most of the total execution time required for BCM. In this paper, effective implementations on modern vector and scalar processors are investigated. NEC SX-9 and Intel Nehalem-EX are the latest vector and scalar processors. Those processors have much higher peak performances than their previous-generation processors. However, their memory bandwidth improvement cannot catch up with the performance improvement of processors. This is the so-called memory wall problem. In our paper, we discuss optimization techniques for implementation of the SOR method based on architectural characteristics of these modern processors, and evaluate their effects on the sustained performances of these processors for BCM.</description><subject>Algorithms</subject><subject>Building-cube method</subject><subject>Computational fluid dynamics</subject><subject>Computer simulation</subject><subject>Mathematical analysis</subject><subject>Mathematical models</subject><subject>Processors</subject><subject>Scalars</subject><subject>SOR method</subject><subject>Vector and scalar processing</subject><subject>Vectors (mathematics)</subject><issn>0045-7930</issn><issn>1879-0747</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2011</creationdate><recordtype>article</recordtype><recordid>eNqFkE1PwzAMhiMEEmPwG8gNLi3OR5v2OE18SZOG-DhHaeqITm0zkg6Jf0-mIY5wsC1br1_LDyGXDHIGrLzZ5NYPW9fvujbnsJ_yHLg8IjNWqToDJdUxmQHIIlO1gFNyFuMGUi-4nJHFEwbnw2BGi9Q7-rJ-pgNO776N1I908C2GkX6inXygZmxptKY3gW6DtxijD_GcnDjTR7z4qXPydnf7unzIVuv7x-VilVkpyymrsQRTpqiksU41NUvZFBKgkswwIQrDDaIUdcNRMctZowRrnLO8lKIFMSdXB990-mOHcdJDFy32vRnR76KuVAGCl1WdlNd_KpmCWlWlUEWSqoPUBh9jQKe3oRtM-NIM9B6v3uhfvHqPVzOuE960uThsYvr5s8Ogo-0wUWy7kGjp1nf_enwDHK6HIw</recordid><startdate>20110601</startdate><enddate>20110601</enddate><creator>Soga, Takashi</creator><creator>Musa, Akihiro</creator><creator>Okabe, Koki</creator><creator>Komatsu, Kazuhiko</creator><creator>Egawa, Ryusuke</creator><creator>Takizawa, Hiroyuki</creator><creator>Kobayashi, Hiroaki</creator><creator>Takahashi, Shun</creator><creator>Sasaki, Daisuke</creator><creator>Nakahashi, Kazuhiro</creator><general>Elsevier Ltd</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7TB</scope><scope>7U5</scope><scope>8FD</scope><scope>FR3</scope><scope>H8D</scope><scope>JQ2</scope><scope>KR7</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>7UA</scope><scope>C1K</scope><scope>F1W</scope><scope>H96</scope><scope>L.G</scope></search><sort><creationdate>20110601</creationdate><title>Performance of SOR methods on modern vector and scalar processors</title><author>Soga, Takashi ; Musa, Akihiro ; Okabe, Koki ; Komatsu, Kazuhiko ; Egawa, Ryusuke ; Takizawa, Hiroyuki ; Kobayashi, Hiroaki ; Takahashi, Shun ; Sasaki, Daisuke ; Nakahashi, Kazuhiro</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c446t-9e60a660a84acf7b91cf7a5400841a1335a2aee439b2e71c21b731bffc2643d03</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2011</creationdate><topic>Algorithms</topic><topic>Building-cube method</topic><topic>Computational fluid dynamics</topic><topic>Computer simulation</topic><topic>Mathematical analysis</topic><topic>Mathematical models</topic><topic>Processors</topic><topic>Scalars</topic><topic>SOR method</topic><topic>Vector and scalar processing</topic><topic>Vectors (mathematics)</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Soga, Takashi</creatorcontrib><creatorcontrib>Musa, Akihiro</creatorcontrib><creatorcontrib>Okabe, Koki</creatorcontrib><creatorcontrib>Komatsu, Kazuhiko</creatorcontrib><creatorcontrib>Egawa, Ryusuke</creatorcontrib><creatorcontrib>Takizawa, Hiroyuki</creatorcontrib><creatorcontrib>Kobayashi, Hiroaki</creatorcontrib><creatorcontrib>Takahashi, Shun</creatorcontrib><creatorcontrib>Sasaki, Daisuke</creatorcontrib><creatorcontrib>Nakahashi, Kazuhiro</creatorcontrib><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Mechanical & Transportation Engineering Abstracts</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>Technology Research Database</collection><collection>Engineering Research Database</collection><collection>Aerospace Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Civil Engineering Abstracts</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>Water Resources Abstracts</collection><collection>Environmental Sciences and Pollution Management</collection><collection>ASFA: Aquatic Sciences and Fisheries Abstracts</collection><collection>Aquatic Science & Fisheries Abstracts (ASFA) 2: Ocean Technology, Policy & Non-Living Resources</collection><collection>Aquatic Science & Fisheries Abstracts (ASFA) Professional</collection><jtitle>Computers & fluids</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Soga, Takashi</au><au>Musa, Akihiro</au><au>Okabe, Koki</au><au>Komatsu, Kazuhiko</au><au>Egawa, Ryusuke</au><au>Takizawa, Hiroyuki</au><au>Kobayashi, Hiroaki</au><au>Takahashi, Shun</au><au>Sasaki, Daisuke</au><au>Nakahashi, Kazuhiro</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Performance of SOR methods on modern vector and scalar processors</atitle><jtitle>Computers & fluids</jtitle><date>2011-06-01</date><risdate>2011</risdate><volume>45</volume><issue>1</issue><spage>215</spage><epage>221</epage><pages>215-221</pages><issn>0045-7930</issn><eissn>1879-0747</eissn><abstract>The building-cube method (BCM) is a new generation algorithm for CFD simulations. The basic idea of BCM is to simplify the algorithm in all stages of flow computation to achieve large-scale simulations. Calculation of a pressure field using the Successive Over Relaxation (SOR) method consumes most of the total execution time required for BCM. In this paper, effective implementations on modern vector and scalar processors are investigated. NEC SX-9 and Intel Nehalem-EX are the latest vector and scalar processors. Those processors have much higher peak performances than their previous-generation processors. However, their memory bandwidth improvement cannot catch up with the performance improvement of processors. This is the so-called memory wall problem. In our paper, we discuss optimization techniques for implementation of the SOR method based on architectural characteristics of these modern processors, and evaluate their effects on the sustained performances of these processors for BCM.</abstract><pub>Elsevier Ltd</pub><doi>10.1016/j.compfluid.2010.12.024</doi><tpages>7</tpages></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0045-7930 |
| ispartof | Computers & fluids, 2011-06, Vol.45 (1), p.215-221 |
| issn | 0045-7930 1879-0747 |
| language | eng |
| recordid | cdi_proquest_miscellaneous_875032689 |
| source | Elsevier ScienceDirect Journals Complete |
| subjects | Algorithms Building-cube method Computational fluid dynamics Computer simulation Mathematical analysis Mathematical models Processors Scalars SOR method Vector and scalar processing Vectors (mathematics) |
| title | Performance of SOR methods on modern vector and scalar processors |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-09T08%3A07%3A40IST&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=Performance%20of%20SOR%20methods%20on%20modern%20vector%20and%20scalar%20processors&rft.jtitle=Computers%20&%20fluids&rft.au=Soga,%20Takashi&rft.date=2011-06-01&rft.volume=45&rft.issue=1&rft.spage=215&rft.epage=221&rft.pages=215-221&rft.issn=0045-7930&rft.eissn=1879-0747&rft_id=info:doi/10.1016/j.compfluid.2010.12.024&rft_dat=%3Cproquest_cross%3E875032689%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=1709786375&rft_id=info:pmid/&rft_els_id=S0045793010003804&rfr_iscdi=true |