Exploiting locality: a flexible DSM approach

No single coherence strategy suits all applications well. Many promising adaptive protocols and coherence predictors, capable of dynamically modifying the coherence strategy, have been suggested over the years. While most dynamic detection schemes rely on plentiful of dedicated hardware, the customi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Zeffer, H., Radovic, Z., Hagersten, E.
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page 10 pp.
container_title
container_volume
creator Zeffer, H.
Radovic, Z.
Hagersten, E.
description No single coherence strategy suits all applications well. Many promising adaptive protocols and coherence predictors, capable of dynamically modifying the coherence strategy, have been suggested over the years. While most dynamic detection schemes rely on plentiful of dedicated hardware, the customization technique suggested in this paper requires no extra hardware support for its per-application coherence strategy. Instead, each application is profiled using a low-overhead profiling tool. The appropriate coherence flag setting, suggested by the profiling, is specified when the application is launched. We have compared the performance of a hardware DSM (Sun WildFire) to a software DSM (distributed shared memory) built with identical interconnect hardware and coherence strategy. With no support for flexibility, the software DSM runs on average 45 percent slower than the hardware DSM on the 12 studied applications, while the flexibility can get the software DSM within 11 percent. Our all-software system outperforms the hardware DSM on four applications
doi_str_mv 10.1109/IPDPS.2006.1639273
format Conference Proceeding
fullrecord <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_1639273</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>1639273</ieee_id><sourcerecordid>1639273</sourcerecordid><originalsourceid>FETCH-LOGICAL-i90t-d4d6a0f751c5d843d27758fb238ddca384957ffde469d69bbb2cd3073d3f25513</originalsourceid><addsrcrecordid>eNotj8tKAzEUQAMqWGt_QDf5AGe8yc1j4k7aqoWKhXZf8tRIdIaZWbR_b8GezdkdOITcMagZA_O42iw225oDqJopNFzjBblhggsBIIW6JBMmESoOWl6T2TB8wwk00hickIfloSttHvPvJy2ttyWPxydqaSrxkF2JdLF9p7br-tb6r1tylWwZ4uzsKdm9LHfzt2r98bqaP6-rbGCsggjKQtKSeRkagYFrLZvkODYheIuNMFKnFKJQJijjnOM-IGgMmLiUDKfk_j-bY4z7rs8_tj_uz2_4BzqyQe0</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Exploiting locality: a flexible DSM approach</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Zeffer, H. ; Radovic, Z. ; Hagersten, E.</creator><creatorcontrib>Zeffer, H. ; Radovic, Z. ; Hagersten, E.</creatorcontrib><description>No single coherence strategy suits all applications well. Many promising adaptive protocols and coherence predictors, capable of dynamically modifying the coherence strategy, have been suggested over the years. While most dynamic detection schemes rely on plentiful of dedicated hardware, the customization technique suggested in this paper requires no extra hardware support for its per-application coherence strategy. Instead, each application is profiled using a low-overhead profiling tool. The appropriate coherence flag setting, suggested by the profiling, is specified when the application is launched. We have compared the performance of a hardware DSM (Sun WildFire) to a software DSM (distributed shared memory) built with identical interconnect hardware and coherence strategy. With no support for flexibility, the software DSM runs on average 45 percent slower than the hardware DSM on the 12 studied applications, while the flexibility can get the software DSM within 11 percent. Our all-software system outperforms the hardware DSM on four applications</description><identifier>ISSN: 1530-2075</identifier><identifier>ISBN: 1424400546</identifier><identifier>ISBN: 9781424400546</identifier><identifier>DOI: 10.1109/IPDPS.2006.1639273</identifier><language>eng</language><publisher>IEEE</publisher><subject>Application software ; Coherence ; Costs ; Hardware ; Information technology ; Proposals ; Protocols ; Software performance ; Sun ; Yarn</subject><ispartof>Proceedings 20th IEEE International Parallel &amp; Distributed Processing Symposium, 2006, p.10 pp.</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/1639273$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,780,784,789,790,2056,4047,4048,27923,54918</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/1639273$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Zeffer, H.</creatorcontrib><creatorcontrib>Radovic, Z.</creatorcontrib><creatorcontrib>Hagersten, E.</creatorcontrib><title>Exploiting locality: a flexible DSM approach</title><title>Proceedings 20th IEEE International Parallel &amp; Distributed Processing Symposium</title><addtitle>IPDPS</addtitle><description>No single coherence strategy suits all applications well. Many promising adaptive protocols and coherence predictors, capable of dynamically modifying the coherence strategy, have been suggested over the years. While most dynamic detection schemes rely on plentiful of dedicated hardware, the customization technique suggested in this paper requires no extra hardware support for its per-application coherence strategy. Instead, each application is profiled using a low-overhead profiling tool. The appropriate coherence flag setting, suggested by the profiling, is specified when the application is launched. We have compared the performance of a hardware DSM (Sun WildFire) to a software DSM (distributed shared memory) built with identical interconnect hardware and coherence strategy. With no support for flexibility, the software DSM runs on average 45 percent slower than the hardware DSM on the 12 studied applications, while the flexibility can get the software DSM within 11 percent. Our all-software system outperforms the hardware DSM on four applications</description><subject>Application software</subject><subject>Coherence</subject><subject>Costs</subject><subject>Hardware</subject><subject>Information technology</subject><subject>Proposals</subject><subject>Protocols</subject><subject>Software performance</subject><subject>Sun</subject><subject>Yarn</subject><issn>1530-2075</issn><isbn>1424400546</isbn><isbn>9781424400546</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2006</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNotj8tKAzEUQAMqWGt_QDf5AGe8yc1j4k7aqoWKhXZf8tRIdIaZWbR_b8GezdkdOITcMagZA_O42iw225oDqJopNFzjBblhggsBIIW6JBMmESoOWl6T2TB8wwk00hickIfloSttHvPvJy2ttyWPxydqaSrxkF2JdLF9p7br-tb6r1tylWwZ4uzsKdm9LHfzt2r98bqaP6-rbGCsggjKQtKSeRkagYFrLZvkODYheIuNMFKnFKJQJijjnOM-IGgMmLiUDKfk_j-bY4z7rs8_tj_uz2_4BzqyQe0</recordid><startdate>2006</startdate><enddate>2006</enddate><creator>Zeffer, H.</creator><creator>Radovic, Z.</creator><creator>Hagersten, E.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>2006</creationdate><title>Exploiting locality: a flexible DSM approach</title><author>Zeffer, H. ; Radovic, Z. ; Hagersten, E.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i90t-d4d6a0f751c5d843d27758fb238ddca384957ffde469d69bbb2cd3073d3f25513</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2006</creationdate><topic>Application software</topic><topic>Coherence</topic><topic>Costs</topic><topic>Hardware</topic><topic>Information technology</topic><topic>Proposals</topic><topic>Protocols</topic><topic>Software performance</topic><topic>Sun</topic><topic>Yarn</topic><toplevel>online_resources</toplevel><creatorcontrib>Zeffer, H.</creatorcontrib><creatorcontrib>Radovic, Z.</creatorcontrib><creatorcontrib>Hagersten, E.</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></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Zeffer, H.</au><au>Radovic, Z.</au><au>Hagersten, E.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Exploiting locality: a flexible DSM approach</atitle><btitle>Proceedings 20th IEEE International Parallel &amp; Distributed Processing Symposium</btitle><stitle>IPDPS</stitle><date>2006</date><risdate>2006</risdate><spage>10 pp.</spage><pages>10 pp.-</pages><issn>1530-2075</issn><isbn>1424400546</isbn><isbn>9781424400546</isbn><abstract>No single coherence strategy suits all applications well. Many promising adaptive protocols and coherence predictors, capable of dynamically modifying the coherence strategy, have been suggested over the years. While most dynamic detection schemes rely on plentiful of dedicated hardware, the customization technique suggested in this paper requires no extra hardware support for its per-application coherence strategy. Instead, each application is profiled using a low-overhead profiling tool. The appropriate coherence flag setting, suggested by the profiling, is specified when the application is launched. We have compared the performance of a hardware DSM (Sun WildFire) to a software DSM (distributed shared memory) built with identical interconnect hardware and coherence strategy. With no support for flexibility, the software DSM runs on average 45 percent slower than the hardware DSM on the 12 studied applications, while the flexibility can get the software DSM within 11 percent. Our all-software system outperforms the hardware DSM on four applications</abstract><pub>IEEE</pub><doi>10.1109/IPDPS.2006.1639273</doi></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 1530-2075
ispartof Proceedings 20th IEEE International Parallel & Distributed Processing Symposium, 2006, p.10 pp.
issn 1530-2075
language eng
recordid cdi_ieee_primary_1639273
source IEEE Electronic Library (IEL) Conference Proceedings
subjects Application software
Coherence
Costs
Hardware
Information technology
Proposals
Protocols
Software performance
Sun
Yarn
title Exploiting locality: a flexible DSM approach
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-13T13%3A34%3A56IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Exploiting%20locality:%20a%20flexible%20DSM%20approach&rft.btitle=Proceedings%2020th%20IEEE%20International%20Parallel%20&%20Distributed%20Processing%20Symposium&rft.au=Zeffer,%20H.&rft.date=2006&rft.spage=10%20pp.&rft.pages=10%20pp.-&rft.issn=1530-2075&rft.isbn=1424400546&rft.isbn_list=9781424400546&rft_id=info:doi/10.1109/IPDPS.2006.1639273&rft_dat=%3Cieee_6IE%3E1639273%3C/ieee_6IE%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=1639273&rfr_iscdi=true