Characterizing and evaluating desktop grids: an empirical study
Summary form only given. Desktop resources are attractive for running compute-intensive distributed applications. Several systems that aggregate these resources in desktop grids have been developed. While these systems have been successfully used for many high throughput applications there has been...
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 | |
|---|---|
| container_issue | |
| container_start_page | 26 |
| container_title | |
| container_volume | |
| creator | Kondo, D. Taufer, M. Brooks, C.L. Casanova, H. Chien, A.A. |
| description | Summary form only given. Desktop resources are attractive for running compute-intensive distributed applications. Several systems that aggregate these resources in desktop grids have been developed. While these systems have been successfully used for many high throughput applications there has been little insight into the detailed temporal structure of CPU availability of desktop grid resources. Yet, this structure is critical to characterize the utility of desktop grid platforms for both task parallel and even data parallel applications. We address the following questions: (i) What are the temporal characteristics of desktop CPU availability in an enterprise setting? (ii) How do these characteristics affect the utility of desktop grids? (iii) Based on these characteristics, can we construct a model of server "equivalents" for the desktop grids, which can be used to predict application performance? We present measurements of an enterprise desktop grid with over 220 hosts running the Entropia commercial desktop grid software. We utilize these measurements to characterize CPU availability and develop a performance model for desktop grid applications for various task granularities, showing that there is an optimal task size. We then use a cluster equivalence metric to quantify the utility of the desktop grid relative to that of a dedicated cluster. |
| doi_str_mv | 10.1109/IPDPS.2004.1302936 |
| format | Conference Proceeding |
| fullrecord | <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_1302936</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>1302936</ieee_id><sourcerecordid>1302936</sourcerecordid><originalsourceid>FETCH-LOGICAL-i646-6dfacb5c1e5a0242f88215b82bfbbdcecc6056148a9012c19a98274eaa2738683</originalsourceid><addsrcrecordid>eNotj81KAzEUhQMiKHVeQDfzAlOTm59J3IiMWgsFC3Zf7iR3anRah2Qq1Ke3Ys_m8HHgg8PYteBTIbi7nS8fl29T4FxNheTgpDljhastr43TICS4C1bk_MGPUUqrWl-y--YdE_qRUvyJu02Ju1DSN_Z7HP8wUP4cv4Zyk2LId8e1pO0QU_TYl3nch8MVO--wz1ScesJWz0-r5qVavM7mzcOiikaZyoQOfau9II0cFHTWgtCthbZr2-DJe8O1Ecqi4wK8cOgs1IoQoZbWWDlhN__aSETrIcUtpsP69FL-AodGSDQ</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>Characterizing and evaluating desktop grids: an empirical study</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Kondo, D. ; Taufer, M. ; Brooks, C.L. ; Casanova, H. ; Chien, A.A.</creator><creatorcontrib>Kondo, D. ; Taufer, M. ; Brooks, C.L. ; Casanova, H. ; Chien, A.A.</creatorcontrib><description>Summary form only given. Desktop resources are attractive for running compute-intensive distributed applications. Several systems that aggregate these resources in desktop grids have been developed. While these systems have been successfully used for many high throughput applications there has been little insight into the detailed temporal structure of CPU availability of desktop grid resources. Yet, this structure is critical to characterize the utility of desktop grid platforms for both task parallel and even data parallel applications. We address the following questions: (i) What are the temporal characteristics of desktop CPU availability in an enterprise setting? (ii) How do these characteristics affect the utility of desktop grids? (iii) Based on these characteristics, can we construct a model of server "equivalents" for the desktop grids, which can be used to predict application performance? We present measurements of an enterprise desktop grid with over 220 hosts running the Entropia commercial desktop grid software. We utilize these measurements to characterize CPU availability and develop a performance model for desktop grid applications for various task granularities, showing that there is an optimal task size. We then use a cluster equivalence metric to quantify the utility of the desktop grid relative to that of a dedicated cluster.</description><identifier>ISBN: 9780769521329</identifier><identifier>ISBN: 0769521320</identifier><identifier>DOI: 10.1109/IPDPS.2004.1302936</identifier><language>eng</language><publisher>IEEE</publisher><subject>Aggregates ; Application software ; Availability ; Biology computing ; Computer applications ; Computer science ; Distributed computing ; Predictive models ; Supercomputers ; Throughput</subject><ispartof>18th International Parallel and Distributed Processing Symposium, 2004. Proceedings, 2004, p.26</ispartof><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/1302936$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,776,780,785,786,2052,4036,4037,27902,54895</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/1302936$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Kondo, D.</creatorcontrib><creatorcontrib>Taufer, M.</creatorcontrib><creatorcontrib>Brooks, C.L.</creatorcontrib><creatorcontrib>Casanova, H.</creatorcontrib><creatorcontrib>Chien, A.A.</creatorcontrib><title>Characterizing and evaluating desktop grids: an empirical study</title><title>18th International Parallel and Distributed Processing Symposium, 2004. Proceedings</title><addtitle>IPDPS</addtitle><description>Summary form only given. Desktop resources are attractive for running compute-intensive distributed applications. Several systems that aggregate these resources in desktop grids have been developed. While these systems have been successfully used for many high throughput applications there has been little insight into the detailed temporal structure of CPU availability of desktop grid resources. Yet, this structure is critical to characterize the utility of desktop grid platforms for both task parallel and even data parallel applications. We address the following questions: (i) What are the temporal characteristics of desktop CPU availability in an enterprise setting? (ii) How do these characteristics affect the utility of desktop grids? (iii) Based on these characteristics, can we construct a model of server "equivalents" for the desktop grids, which can be used to predict application performance? We present measurements of an enterprise desktop grid with over 220 hosts running the Entropia commercial desktop grid software. We utilize these measurements to characterize CPU availability and develop a performance model for desktop grid applications for various task granularities, showing that there is an optimal task size. We then use a cluster equivalence metric to quantify the utility of the desktop grid relative to that of a dedicated cluster.</description><subject>Aggregates</subject><subject>Application software</subject><subject>Availability</subject><subject>Biology computing</subject><subject>Computer applications</subject><subject>Computer science</subject><subject>Distributed computing</subject><subject>Predictive models</subject><subject>Supercomputers</subject><subject>Throughput</subject><isbn>9780769521329</isbn><isbn>0769521320</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2004</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNotj81KAzEUhQMiKHVeQDfzAlOTm59J3IiMWgsFC3Zf7iR3anRah2Qq1Ke3Ys_m8HHgg8PYteBTIbi7nS8fl29T4FxNheTgpDljhastr43TICS4C1bk_MGPUUqrWl-y--YdE_qRUvyJu02Ju1DSN_Z7HP8wUP4cv4Zyk2LId8e1pO0QU_TYl3nch8MVO--wz1ScesJWz0-r5qVavM7mzcOiikaZyoQOfau9II0cFHTWgtCthbZr2-DJe8O1Ecqi4wK8cOgs1IoQoZbWWDlhN__aSETrIcUtpsP69FL-AodGSDQ</recordid><startdate>2004</startdate><enddate>2004</enddate><creator>Kondo, D.</creator><creator>Taufer, M.</creator><creator>Brooks, C.L.</creator><creator>Casanova, H.</creator><creator>Chien, A.A.</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>2004</creationdate><title>Characterizing and evaluating desktop grids: an empirical study</title><author>Kondo, D. ; Taufer, M. ; Brooks, C.L. ; Casanova, H. ; Chien, A.A.</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i646-6dfacb5c1e5a0242f88215b82bfbbdcecc6056148a9012c19a98274eaa2738683</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2004</creationdate><topic>Aggregates</topic><topic>Application software</topic><topic>Availability</topic><topic>Biology computing</topic><topic>Computer applications</topic><topic>Computer science</topic><topic>Distributed computing</topic><topic>Predictive models</topic><topic>Supercomputers</topic><topic>Throughput</topic><toplevel>online_resources</toplevel><creatorcontrib>Kondo, D.</creatorcontrib><creatorcontrib>Taufer, M.</creatorcontrib><creatorcontrib>Brooks, C.L.</creatorcontrib><creatorcontrib>Casanova, H.</creatorcontrib><creatorcontrib>Chien, A.A.</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>Kondo, D.</au><au>Taufer, M.</au><au>Brooks, C.L.</au><au>Casanova, H.</au><au>Chien, A.A.</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Characterizing and evaluating desktop grids: an empirical study</atitle><btitle>18th International Parallel and Distributed Processing Symposium, 2004. Proceedings</btitle><stitle>IPDPS</stitle><date>2004</date><risdate>2004</risdate><spage>26</spage><pages>26-</pages><isbn>9780769521329</isbn><isbn>0769521320</isbn><abstract>Summary form only given. Desktop resources are attractive for running compute-intensive distributed applications. Several systems that aggregate these resources in desktop grids have been developed. While these systems have been successfully used for many high throughput applications there has been little insight into the detailed temporal structure of CPU availability of desktop grid resources. Yet, this structure is critical to characterize the utility of desktop grid platforms for both task parallel and even data parallel applications. We address the following questions: (i) What are the temporal characteristics of desktop CPU availability in an enterprise setting? (ii) How do these characteristics affect the utility of desktop grids? (iii) Based on these characteristics, can we construct a model of server "equivalents" for the desktop grids, which can be used to predict application performance? We present measurements of an enterprise desktop grid with over 220 hosts running the Entropia commercial desktop grid software. We utilize these measurements to characterize CPU availability and develop a performance model for desktop grid applications for various task granularities, showing that there is an optimal task size. We then use a cluster equivalence metric to quantify the utility of the desktop grid relative to that of a dedicated cluster.</abstract><pub>IEEE</pub><doi>10.1109/IPDPS.2004.1302936</doi><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext_linktorsrc |
| identifier | ISBN: 9780769521329 |
| ispartof | 18th International Parallel and Distributed Processing Symposium, 2004. Proceedings, 2004, p.26 |
| issn | |
| language | eng |
| recordid | cdi_ieee_primary_1302936 |
| source | IEEE Electronic Library (IEL) Conference Proceedings |
| subjects | Aggregates Application software Availability Biology computing Computer applications Computer science Distributed computing Predictive models Supercomputers Throughput |
| title | Characterizing and evaluating desktop grids: an empirical study |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-31T15%3A26%3A29IST&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=Characterizing%20and%20evaluating%20desktop%20grids:%20an%20empirical%20study&rft.btitle=18th%20International%20Parallel%20and%20Distributed%20Processing%20Symposium,%202004.%20Proceedings&rft.au=Kondo,%20D.&rft.date=2004&rft.spage=26&rft.pages=26-&rft.isbn=9780769521329&rft.isbn_list=0769521320&rft_id=info:doi/10.1109/IPDPS.2004.1302936&rft_dat=%3Cieee_6IE%3E1302936%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=1302936&rfr_iscdi=true |