Heterogeneous Networks of Workstations

Parallel systems built from commodity CPUs and networking devices (Networks of Workstations or NoW) are easily and economically constructed. Rapid technological change means that as elements are added to or replaced in such a system, it will inevitably become heterogeneous. This paper examines issue...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Baek, SunHo, Lee, KyuHo, Kim, JunSeong, Morris, John
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 439
container_issue
container_start_page 426
container_title
container_volume
creator Baek, SunHo
Lee, KyuHo
Kim, JunSeong
Morris, John
description Parallel systems built from commodity CPUs and networking devices (Networks of Workstations or NoW) are easily and economically constructed. Rapid technological change means that as elements are added to or replaced in such a system, it will inevitably become heterogeneous. This paper examines issues in effectively using all the processing power of a heterogeneous NoW system. In particular, it examines the use of two different parallel programming models: MPI, a popular message passing system and Cilk, an extension to C with dataflow semantics. We consider both the performance of the two systems for a set of simple benchmarks representative of real problems and the ease of programming these systems. We show that, in many cases, Cilk both performs better than MPI and, additionally, that it requires less programming effort to write programs that exploit the full power of a heterogeneous system.
doi_str_mv 10.1007/978-3-540-30102-8_36
format Conference Proceeding
fullrecord <record><control><sourceid>pascalfrancis_sprin</sourceid><recordid>TN_cdi_pascalfrancis_primary_16177809</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>16177809</sourcerecordid><originalsourceid>FETCH-LOGICAL-p228t-9d8ac78f649a0ed0b7b8f331b3d7d9a3359975e9af1e0686d7e6342929b9ad0a3</originalsourceid><addsrcrecordid>eNotkE1PwzAMhsOXRBn9Bxx6gVvAidsmPqJpMKQJLiC4RWmbTmWjmZoixL8n3fDF1utHlvwwdiXgVgCoO1KaIy9y4AgCJNcGyyN2gTHZBx_HLBGlEBwxpxOWRn7aSQRAfcqSSElOKsdzlobwCVORxlwk7GbpRjf4teud_w7Zsxt__LAJmW-z92kY7dj5Plyys9Zug0v_-4y9PSxe50u-enl8mt-v-E5KPXJqtK2VbsucLLgGKlXpFlFU2KiGLGJBpApHthUOSl02ypWYS5JUkW3A4oxdH-7ubKjtth1sX3fB7Ibuyw6_Jj6plAaKnDxwIa76tRtM5f0mGAFmMmaiAYMmOjB7QWYyhn-1TFiM</addsrcrecordid><sourcetype>Index Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Heterogeneous Networks of Workstations</title><source>Springer Books</source><creator>Baek, SunHo ; Lee, KyuHo ; Kim, JunSeong ; Morris, John</creator><contributor>Xue, Jingling ; Yew, Pen-Chung</contributor><creatorcontrib>Baek, SunHo ; Lee, KyuHo ; Kim, JunSeong ; Morris, John ; Xue, Jingling ; Yew, Pen-Chung</creatorcontrib><description>Parallel systems built from commodity CPUs and networking devices (Networks of Workstations or NoW) are easily and economically constructed. Rapid technological change means that as elements are added to or replaced in such a system, it will inevitably become heterogeneous. This paper examines issues in effectively using all the processing power of a heterogeneous NoW system. In particular, it examines the use of two different parallel programming models: MPI, a popular message passing system and Cilk, an extension to C with dataflow semantics. We consider both the performance of the two systems for a set of simple benchmarks representative of real problems and the ease of programming these systems. We show that, in many cases, Cilk both performs better than MPI and, additionally, that it requires less programming effort to write programs that exploit the full power of a heterogeneous system.</description><identifier>ISSN: 0302-9743</identifier><identifier>ISBN: 9783540230038</identifier><identifier>ISBN: 3540230033</identifier><identifier>EISSN: 1611-3349</identifier><identifier>EISBN: 354030102X</identifier><identifier>EISBN: 9783540301028</identifier><identifier>DOI: 10.1007/978-3-540-30102-8_36</identifier><language>eng</language><publisher>Berlin, Heidelberg: Springer Berlin Heidelberg</publisher><subject>Applied sciences ; Basic Speedup ; Computers, microcomputers ; Electronics ; Exact sciences and technology ; Good Tour ; Hardware ; Heterogeneous Network ; Integrated circuits ; Integrated circuits by function (including memories and processors) ; Runtime System ; Semiconductor electronics. Microelectronics. Optoelectronics. Solid state devices ; Travelling Salesman Problem</subject><ispartof>Advances in Computer Systems Architecture, 2004, p.426-439</ispartof><rights>Springer-Verlag Berlin Heidelberg 2004</rights><rights>2004 INIST-CNRS</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/978-3-540-30102-8_36$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/978-3-540-30102-8_36$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>309,310,775,776,780,785,786,789,4035,4036,27904,38234,41421,42490</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&amp;idt=16177809$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><contributor>Xue, Jingling</contributor><contributor>Yew, Pen-Chung</contributor><creatorcontrib>Baek, SunHo</creatorcontrib><creatorcontrib>Lee, KyuHo</creatorcontrib><creatorcontrib>Kim, JunSeong</creatorcontrib><creatorcontrib>Morris, John</creatorcontrib><title>Heterogeneous Networks of Workstations</title><title>Advances in Computer Systems Architecture</title><description>Parallel systems built from commodity CPUs and networking devices (Networks of Workstations or NoW) are easily and economically constructed. Rapid technological change means that as elements are added to or replaced in such a system, it will inevitably become heterogeneous. This paper examines issues in effectively using all the processing power of a heterogeneous NoW system. In particular, it examines the use of two different parallel programming models: MPI, a popular message passing system and Cilk, an extension to C with dataflow semantics. We consider both the performance of the two systems for a set of simple benchmarks representative of real problems and the ease of programming these systems. We show that, in many cases, Cilk both performs better than MPI and, additionally, that it requires less programming effort to write programs that exploit the full power of a heterogeneous system.</description><subject>Applied sciences</subject><subject>Basic Speedup</subject><subject>Computers, microcomputers</subject><subject>Electronics</subject><subject>Exact sciences and technology</subject><subject>Good Tour</subject><subject>Hardware</subject><subject>Heterogeneous Network</subject><subject>Integrated circuits</subject><subject>Integrated circuits by function (including memories and processors)</subject><subject>Runtime System</subject><subject>Semiconductor electronics. Microelectronics. Optoelectronics. Solid state devices</subject><subject>Travelling Salesman Problem</subject><issn>0302-9743</issn><issn>1611-3349</issn><isbn>9783540230038</isbn><isbn>3540230033</isbn><isbn>354030102X</isbn><isbn>9783540301028</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2004</creationdate><recordtype>conference_proceeding</recordtype><recordid>eNotkE1PwzAMhsOXRBn9Bxx6gVvAidsmPqJpMKQJLiC4RWmbTmWjmZoixL8n3fDF1utHlvwwdiXgVgCoO1KaIy9y4AgCJNcGyyN2gTHZBx_HLBGlEBwxpxOWRn7aSQRAfcqSSElOKsdzlobwCVORxlwk7GbpRjf4teud_w7Zsxt__LAJmW-z92kY7dj5Plyys9Zug0v_-4y9PSxe50u-enl8mt-v-E5KPXJqtK2VbsucLLgGKlXpFlFU2KiGLGJBpApHthUOSl02ypWYS5JUkW3A4oxdH-7ubKjtth1sX3fB7Ibuyw6_Jj6plAaKnDxwIa76tRtM5f0mGAFmMmaiAYMmOjB7QWYyhn-1TFiM</recordid><startdate>2004</startdate><enddate>2004</enddate><creator>Baek, SunHo</creator><creator>Lee, KyuHo</creator><creator>Kim, JunSeong</creator><creator>Morris, John</creator><general>Springer Berlin Heidelberg</general><general>Springer</general><scope>IQODW</scope></search><sort><creationdate>2004</creationdate><title>Heterogeneous Networks of Workstations</title><author>Baek, SunHo ; Lee, KyuHo ; Kim, JunSeong ; Morris, John</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-p228t-9d8ac78f649a0ed0b7b8f331b3d7d9a3359975e9af1e0686d7e6342929b9ad0a3</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2004</creationdate><topic>Applied sciences</topic><topic>Basic Speedup</topic><topic>Computers, microcomputers</topic><topic>Electronics</topic><topic>Exact sciences and technology</topic><topic>Good Tour</topic><topic>Hardware</topic><topic>Heterogeneous Network</topic><topic>Integrated circuits</topic><topic>Integrated circuits by function (including memories and processors)</topic><topic>Runtime System</topic><topic>Semiconductor electronics. Microelectronics. Optoelectronics. Solid state devices</topic><topic>Travelling Salesman Problem</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Baek, SunHo</creatorcontrib><creatorcontrib>Lee, KyuHo</creatorcontrib><creatorcontrib>Kim, JunSeong</creatorcontrib><creatorcontrib>Morris, John</creatorcontrib><collection>Pascal-Francis</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Baek, SunHo</au><au>Lee, KyuHo</au><au>Kim, JunSeong</au><au>Morris, John</au><au>Xue, Jingling</au><au>Yew, Pen-Chung</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Heterogeneous Networks of Workstations</atitle><btitle>Advances in Computer Systems Architecture</btitle><date>2004</date><risdate>2004</risdate><spage>426</spage><epage>439</epage><pages>426-439</pages><issn>0302-9743</issn><eissn>1611-3349</eissn><isbn>9783540230038</isbn><isbn>3540230033</isbn><eisbn>354030102X</eisbn><eisbn>9783540301028</eisbn><abstract>Parallel systems built from commodity CPUs and networking devices (Networks of Workstations or NoW) are easily and economically constructed. Rapid technological change means that as elements are added to or replaced in such a system, it will inevitably become heterogeneous. This paper examines issues in effectively using all the processing power of a heterogeneous NoW system. In particular, it examines the use of two different parallel programming models: MPI, a popular message passing system and Cilk, an extension to C with dataflow semantics. We consider both the performance of the two systems for a set of simple benchmarks representative of real problems and the ease of programming these systems. We show that, in many cases, Cilk both performs better than MPI and, additionally, that it requires less programming effort to write programs that exploit the full power of a heterogeneous system.</abstract><cop>Berlin, Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><doi>10.1007/978-3-540-30102-8_36</doi><tpages>14</tpages></addata></record>
fulltext fulltext
identifier ISSN: 0302-9743
ispartof Advances in Computer Systems Architecture, 2004, p.426-439
issn 0302-9743
1611-3349
language eng
recordid cdi_pascalfrancis_primary_16177809
source Springer Books
subjects Applied sciences
Basic Speedup
Computers, microcomputers
Electronics
Exact sciences and technology
Good Tour
Hardware
Heterogeneous Network
Integrated circuits
Integrated circuits by function (including memories and processors)
Runtime System
Semiconductor electronics. Microelectronics. Optoelectronics. Solid state devices
Travelling Salesman Problem
title Heterogeneous Networks of Workstations
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-23T01%3A03%3A27IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-pascalfrancis_sprin&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Heterogeneous%20Networks%20of%20Workstations&rft.btitle=Advances%20in%20Computer%20Systems%20Architecture&rft.au=Baek,%20SunHo&rft.date=2004&rft.spage=426&rft.epage=439&rft.pages=426-439&rft.issn=0302-9743&rft.eissn=1611-3349&rft.isbn=9783540230038&rft.isbn_list=3540230033&rft_id=info:doi/10.1007/978-3-540-30102-8_36&rft_dat=%3Cpascalfrancis_sprin%3E16177809%3C/pascalfrancis_sprin%3E%3Curl%3E%3C/url%3E&rft.eisbn=354030102X&rft.eisbn_list=9783540301028&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true