MPI backend for an automatic parallelizing compiler
Many naive parallel processing schemes were not as successful as many researchers thought, because of the heavy cost of communication and synchronization resulting from parallelization. In this paper, we identify the reasons for this poor performance and the compiler requirements for performance imp...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Tagungsbericht |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext bestellen |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 157 |
|---|---|
| container_issue | |
| container_start_page | 152 |
| container_title | |
| container_volume | |
| creator | Kwon, Daesuk Han, Sangyong Kim, Heunghwan |
| description | Many naive parallel processing schemes were not as successful as many researchers thought, because of the heavy cost of communication and synchronization resulting from parallelization. In this paper, we identify the reasons for this poor performance and the compiler requirements for performance improvement. We realized that the parallelizing decisions should be derived from the overhead information. We added this idea to the automatic parallelizing compiler SUIF. We substituted the original backend of SUIF with our backend using MPI, and gave it the capability to validate parallelization decisions based on overhead parameters. This backend converts shared memory-based parallel programs into distributed memory-based parallel programs with MPI function calls without excessive parallelization, which causes performance degradation. |
| doi_str_mv | 10.1109/ISPAN.1999.778932 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>proquest_6IE</sourceid><recordid>TN_cdi_ieee_primary_778932</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>778932</ieee_id><sourcerecordid>27005378</sourcerecordid><originalsourceid>FETCH-LOGICAL-i203t-fd141e26af6c94612672e639b123367bcfdf5f482b98029ecfdc5329ef49c9783</originalsourceid><addsrcrecordid>eNotkMlOwzAURS0GibTwAbDKil2Ch8T2W1YVhUoFKgESu8hxnpHBGciwgK-nIqzuvdLRWVxCLhlNGaNws33erx5TBgCpUhoEPyIRFypPcq7ejsmCKgk55YLpExIxqlWSUQ1nZDEMH5RyKkBGRDzst3Fp7Cc2VezaPjZNbKaxrc3obdyZ3oSAwf_45j22bd35gP05OXUmDHjxn0vyurl9Wd8nu6e77Xq1S_xBPiauYhlDLo2TFjLJuFQcpYCScSGkKq2rXO4yzUvQlAMets3FobgMLCgtluR69nZ9-zXhMBa1HyyGYBpsp6HgitJc_IFXM-gRseh6X5v-u5hPEb8kxFMj</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype><pqid>27005378</pqid></control><display><type>conference_proceeding</type><title>MPI backend for an automatic parallelizing compiler</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Kwon, Daesuk ; Han, Sangyong ; Kim, Heunghwan</creator><creatorcontrib>Kwon, Daesuk ; Han, Sangyong ; Kim, Heunghwan</creatorcontrib><description>Many naive parallel processing schemes were not as successful as many researchers thought, because of the heavy cost of communication and synchronization resulting from parallelization. In this paper, we identify the reasons for this poor performance and the compiler requirements for performance improvement. We realized that the parallelizing decisions should be derived from the overhead information. We added this idea to the automatic parallelizing compiler SUIF. We substituted the original backend of SUIF with our backend using MPI, and gave it the capability to validate parallelization decisions based on overhead parameters. This backend converts shared memory-based parallel programs into distributed memory-based parallel programs with MPI function calls without excessive parallelization, which causes performance degradation.</description><identifier>ISSN: 1087-4089</identifier><identifier>ISBN: 0769502318</identifier><identifier>ISBN: 9780769502311</identifier><identifier>EISSN: 2375-527X</identifier><identifier>DOI: 10.1109/ISPAN.1999.778932</identifier><language>eng</language><publisher>IEEE</publisher><subject>Clocks ; Computer science ; Costs ; Frequency ; High performance computing ; Linux ; Microprocessors ; Multiprocessing systems ; Parallel processing ; Parallel programming</subject><ispartof>Proceedings / International Symposium on Parallel Architectures, Algorithms, and Networks (ISPAN), 1999, p.152-157</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/778932$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2056,4048,4049,27924,54919</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/778932$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Kwon, Daesuk</creatorcontrib><creatorcontrib>Han, Sangyong</creatorcontrib><creatorcontrib>Kim, Heunghwan</creatorcontrib><title>MPI backend for an automatic parallelizing compiler</title><title>Proceedings / International Symposium on Parallel Architectures, Algorithms, and Networks (ISPAN)</title><addtitle>ISPAN</addtitle><description>Many naive parallel processing schemes were not as successful as many researchers thought, because of the heavy cost of communication and synchronization resulting from parallelization. In this paper, we identify the reasons for this poor performance and the compiler requirements for performance improvement. We realized that the parallelizing decisions should be derived from the overhead information. We added this idea to the automatic parallelizing compiler SUIF. We substituted the original backend of SUIF with our backend using MPI, and gave it the capability to validate parallelization decisions based on overhead parameters. This backend converts shared memory-based parallel programs into distributed memory-based parallel programs with MPI function calls without excessive parallelization, which causes performance degradation.</description><subject>Clocks</subject><subject>Computer science</subject><subject>Costs</subject><subject>Frequency</subject><subject>High performance computing</subject><subject>Linux</subject><subject>Microprocessors</subject><subject>Multiprocessing systems</subject><subject>Parallel processing</subject><subject>Parallel programming</subject><issn>1087-4089</issn><issn>2375-527X</issn><isbn>0769502318</isbn><isbn>9780769502311</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>1999</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNotkMlOwzAURS0GibTwAbDKil2Ch8T2W1YVhUoFKgESu8hxnpHBGciwgK-nIqzuvdLRWVxCLhlNGaNws33erx5TBgCpUhoEPyIRFypPcq7ejsmCKgk55YLpExIxqlWSUQ1nZDEMH5RyKkBGRDzst3Fp7Cc2VezaPjZNbKaxrc3obdyZ3oSAwf_45j22bd35gP05OXUmDHjxn0vyurl9Wd8nu6e77Xq1S_xBPiauYhlDLo2TFjLJuFQcpYCScSGkKq2rXO4yzUvQlAMets3FobgMLCgtluR69nZ9-zXhMBa1HyyGYBpsp6HgitJc_IFXM-gRseh6X5v-u5hPEb8kxFMj</recordid><startdate>1999</startdate><enddate>1999</enddate><creator>Kwon, Daesuk</creator><creator>Han, Sangyong</creator><creator>Kim, Heunghwan</creator><general>IEEE</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>1999</creationdate><title>MPI backend for an automatic parallelizing compiler</title><author>Kwon, Daesuk ; Han, Sangyong ; Kim, Heunghwan</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i203t-fd141e26af6c94612672e639b123367bcfdf5f482b98029ecfdc5329ef49c9783</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>1999</creationdate><topic>Clocks</topic><topic>Computer science</topic><topic>Costs</topic><topic>Frequency</topic><topic>High performance computing</topic><topic>Linux</topic><topic>Microprocessors</topic><topic>Multiprocessing systems</topic><topic>Parallel processing</topic><topic>Parallel programming</topic><toplevel>online_resources</toplevel><creatorcontrib>Kwon, Daesuk</creatorcontrib><creatorcontrib>Han, Sangyong</creatorcontrib><creatorcontrib>Kim, Heunghwan</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 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>Kwon, Daesuk</au><au>Han, Sangyong</au><au>Kim, Heunghwan</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>MPI backend for an automatic parallelizing compiler</atitle><btitle>Proceedings / International Symposium on Parallel Architectures, Algorithms, and Networks (ISPAN)</btitle><stitle>ISPAN</stitle><date>1999</date><risdate>1999</risdate><spage>152</spage><epage>157</epage><pages>152-157</pages><issn>1087-4089</issn><eissn>2375-527X</eissn><isbn>0769502318</isbn><isbn>9780769502311</isbn><abstract>Many naive parallel processing schemes were not as successful as many researchers thought, because of the heavy cost of communication and synchronization resulting from parallelization. In this paper, we identify the reasons for this poor performance and the compiler requirements for performance improvement. We realized that the parallelizing decisions should be derived from the overhead information. We added this idea to the automatic parallelizing compiler SUIF. We substituted the original backend of SUIF with our backend using MPI, and gave it the capability to validate parallelization decisions based on overhead parameters. This backend converts shared memory-based parallel programs into distributed memory-based parallel programs with MPI function calls without excessive parallelization, which causes performance degradation.</abstract><pub>IEEE</pub><doi>10.1109/ISPAN.1999.778932</doi><tpages>6</tpages></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISSN: 1087-4089 |
| ispartof | Proceedings / International Symposium on Parallel Architectures, Algorithms, and Networks (ISPAN), 1999, p.152-157 |
| issn | 1087-4089 2375-527X |
| language | eng |
| recordid | cdi_ieee_primary_778932 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Clocks Computer science Costs Frequency High performance computing Linux Microprocessors Multiprocessing systems Parallel processing Parallel programming |
| title | MPI backend for an automatic parallelizing compiler |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-11T15%3A18%3A55IST&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=MPI%20backend%20for%20an%20automatic%20parallelizing%20compiler&rft.btitle=Proceedings%20/%20International%20Symposium%20on%20Parallel%20Architectures,%20Algorithms,%20and%20Networks%20(ISPAN)&rft.au=Kwon,%20Daesuk&rft.date=1999&rft.spage=152&rft.epage=157&rft.pages=152-157&rft.issn=1087-4089&rft.eissn=2375-527X&rft.isbn=0769502318&rft.isbn_list=9780769502311&rft_id=info:doi/10.1109/ISPAN.1999.778932&rft_dat=%3Cproquest_6IE%3E27005378%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=27005378&rft_id=info:pmid/&rft_ieee_id=778932&rfr_iscdi=true |