Data alignment of loop nests without nonlocal communications

In this paper, how to distribute data to different memory modules and how to distribute computations to different processors for execution in a distributed memory parallel computer without nonlocal communications or with minimum nonlocal communications are addressed. Nonlocal communications are much...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Shang, Weijia, Shu, Zhongliang
Format:	Tagungsbericht
Sprache:	eng
Schlagworte:	Concurrent computing Contracts Data engineering Distributed computing Nearest neighbor searches Sufficient conditions Terminology Vectors
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

container_end_page	450
container_issue
container_start_page	439
container_title
container_volume
creator	Shang, Weijia Shu, Zhongliang
description	In this paper, how to distribute data to different memory modules and how to distribute computations to different processors for execution in a distributed memory parallel computer without nonlocal communications or with minimum nonlocal communications are addressed. Nonlocal communications are much more expensive compared to local communications, e.g., nearest neighbor shifts of data. Algorithms are classified to uniform communication algorithms where communication patterns are regular and affine communication algorithms where communication patterns are affine functions of loop index variables. Necessary and sufficient conditions on the existence of mappings without nonlocal communications are presented. If such mappings exist, constraints posed by the mapping without nonlocal communications are constructed and used to guide the selection of mappings with minimum local communications.< >
doi_str_mv	10.1109/ASAP.1994.331782
format	Conference Proceeding
fullrecord	<record><control><sourceid>proquest_6IE</sourceid><recordid>TN_cdi_ieee_primary_331782</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>331782</ieee_id><sourcerecordid>23864129</sourcerecordid><originalsourceid>FETCH-LOGICAL-i118t-f476b4abb76a4305e16227f1c7513cda29b413019e484757432ef163b79273483</originalsourceid><addsrcrecordid>eNotkEtLAzEURgMq2Fb34iord1NzczN5gJtS6wMKCup6yIwZjWSSsckg_nsLdfVtDh-HQ8gFsCUAM9erl9XzEowRS0RQmh-ROdOgpaxB4TGZAZNYSS35KZnn_MUYVyj1jNzc2mKpDf4jDi4WmnoaUhppdLlk-uPLZ5oKjSmG1NlAuzQMU_SdLT7FfEZOehuyO__fBXm727yuH6rt0_3jerWtPIAuVS-UbIVtWyWtQFY7kJyrHjpVA3bvlptWADIwTmihaiWQux4ktsrsLYXGBbk6_I679D3tzZrB586FYKNLU244aimAmz14eQC9c64Zd36wu9_mUAT_AP5rU5Y</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype><pqid>23864129</pqid></control><display><type>conference_proceeding</type><title>Data alignment of loop nests without nonlocal communications</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Shang, Weijia ; Shu, Zhongliang</creator><creatorcontrib>Shang, Weijia ; Shu, Zhongliang</creatorcontrib><description>In this paper, how to distribute data to different memory modules and how to distribute computations to different processors for execution in a distributed memory parallel computer without nonlocal communications or with minimum nonlocal communications are addressed. Nonlocal communications are much more expensive compared to local communications, e.g., nearest neighbor shifts of data. Algorithms are classified to uniform communication algorithms where communication patterns are regular and affine communication algorithms where communication patterns are affine functions of loop index variables. Necessary and sufficient conditions on the existence of mappings without nonlocal communications are presented. If such mappings exist, constraints posed by the mapping without nonlocal communications are constructed and used to guide the selection of mappings with minimum local communications.< ></description><identifier>ISSN: 1063-6862</identifier><identifier>ISBN: 0818665173</identifier><identifier>ISBN: 9780818665172</identifier><identifier>DOI: 10.1109/ASAP.1994.331782</identifier><language>eng</language><publisher>IEEE Comput. Soc. Press</publisher><subject>Concurrent computing ; Contracts ; Data engineering ; Distributed computing ; Nearest neighbor searches ; Sufficient conditions ; Terminology ; Vectors</subject><ispartof>Proceedings of IEEE International Conference on Application Specific Array Processors (ASSAP'94), 1994, p.439-450</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/331782$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2056,4048,4049,27924,54919</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/331782$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Shang, Weijia</creatorcontrib><creatorcontrib>Shu, Zhongliang</creatorcontrib><title>Data alignment of loop nests without nonlocal communications</title><title>Proceedings of IEEE International Conference on Application Specific Array Processors (ASSAP'94)</title><addtitle>ASAP</addtitle><description>In this paper, how to distribute data to different memory modules and how to distribute computations to different processors for execution in a distributed memory parallel computer without nonlocal communications or with minimum nonlocal communications are addressed. Nonlocal communications are much more expensive compared to local communications, e.g., nearest neighbor shifts of data. Algorithms are classified to uniform communication algorithms where communication patterns are regular and affine communication algorithms where communication patterns are affine functions of loop index variables. Necessary and sufficient conditions on the existence of mappings without nonlocal communications are presented. If such mappings exist, constraints posed by the mapping without nonlocal communications are constructed and used to guide the selection of mappings with minimum local communications.< ></description><subject>Concurrent computing</subject><subject>Contracts</subject><subject>Data engineering</subject><subject>Distributed computing</subject><subject>Nearest neighbor searches</subject><subject>Sufficient conditions</subject><subject>Terminology</subject><subject>Vectors</subject><issn>1063-6862</issn><isbn>0818665173</isbn><isbn>9780818665172</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>1994</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNotkEtLAzEURgMq2Fb34iord1NzczN5gJtS6wMKCup6yIwZjWSSsckg_nsLdfVtDh-HQ8gFsCUAM9erl9XzEowRS0RQmh-ROdOgpaxB4TGZAZNYSS35KZnn_MUYVyj1jNzc2mKpDf4jDi4WmnoaUhppdLlk-uPLZ5oKjSmG1NlAuzQMU_SdLT7FfEZOehuyO__fBXm727yuH6rt0_3jerWtPIAuVS-UbIVtWyWtQFY7kJyrHjpVA3bvlptWADIwTmihaiWQux4ktsrsLYXGBbk6_I679D3tzZrB586FYKNLU244aimAmz14eQC9c64Zd36wu9_mUAT_AP5rU5Y</recordid><startdate>1994</startdate><enddate>1994</enddate><creator>Shang, Weijia</creator><creator>Shu, Zhongliang</creator><general>IEEE Comput. Soc. Press</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope><scope>7SC</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>1994</creationdate><title>Data alignment of loop nests without nonlocal communications</title><author>Shang, Weijia ; Shu, Zhongliang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i118t-f476b4abb76a4305e16227f1c7513cda29b413019e484757432ef163b79273483</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>1994</creationdate><topic>Concurrent computing</topic><topic>Contracts</topic><topic>Data engineering</topic><topic>Distributed computing</topic><topic>Nearest neighbor searches</topic><topic>Sufficient conditions</topic><topic>Terminology</topic><topic>Vectors</topic><toplevel>online_resources</toplevel><creatorcontrib>Shang, Weijia</creatorcontrib><creatorcontrib>Shu, Zhongliang</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE/IET Electronic Library (IEL)</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection><collection>Computer and Information Systems 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></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Shang, Weijia</au><au>Shu, Zhongliang</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Data alignment of loop nests without nonlocal communications</atitle><btitle>Proceedings of IEEE International Conference on Application Specific Array Processors (ASSAP'94)</btitle><stitle>ASAP</stitle><date>1994</date><risdate>1994</risdate><spage>439</spage><epage>450</epage><pages>439-450</pages><issn>1063-6862</issn><isbn>0818665173</isbn><isbn>9780818665172</isbn><abstract>In this paper, how to distribute data to different memory modules and how to distribute computations to different processors for execution in a distributed memory parallel computer without nonlocal communications or with minimum nonlocal communications are addressed. Nonlocal communications are much more expensive compared to local communications, e.g., nearest neighbor shifts of data. Algorithms are classified to uniform communication algorithms where communication patterns are regular and affine communication algorithms where communication patterns are affine functions of loop index variables. Necessary and sufficient conditions on the existence of mappings without nonlocal communications are presented. If such mappings exist, constraints posed by the mapping without nonlocal communications are constructed and used to guide the selection of mappings with minimum local communications.< ></abstract><pub>IEEE Comput. Soc. Press</pub><doi>10.1109/ASAP.1994.331782</doi><tpages>12</tpages></addata></record>
fulltext	fulltext_linktorsrc
identifier	ISSN: 1063-6862
ispartof	Proceedings of IEEE International Conference on Application Specific Array Processors (ASSAP'94), 1994, p.439-450
issn	1063-6862
language	eng
recordid	cdi_ieee_primary_331782
source	IEEE Electronic Library (IEL) Conference Proceedings
subjects	Concurrent computing Contracts Data engineering Distributed computing Nearest neighbor searches Sufficient conditions Terminology Vectors
title	Data alignment of loop nests without nonlocal communications
url	https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-10T14%3A06%3A09IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Data%20alignment%20of%20loop%20nests%20without%20nonlocal%20communications&rft.btitle=Proceedings%20of%20IEEE%20International%20Conference%20on%20Application%20Specific%20Array%20Processors%20(ASSAP'94)&rft.au=Shang,%20Weijia&rft.date=1994&rft.spage=439&rft.epage=450&rft.pages=439-450&rft.issn=1063-6862&rft.isbn=0818665173&rft.isbn_list=9780818665172&rft_id=info:doi/10.1109/ASAP.1994.331782&rft_dat=%3Cproquest_6IE%3E23864129%3C/proquest_6IE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=23864129&rft_id=info:pmid/&rft_ieee_id=331782&rfr_iscdi=true