Productive Programming of GPU Clusters with OmpSs
Clusters of GPUs are emerging as a new computational scenario. Programming them requires the use of hybrid models that increase the complexity of the applications, reducing the productivity of programmers. We present the implementation of OmpSs for clusters of GPUs, which supports asynchrony and het...
Uloženo v:
| Vydáno v: | 2012 IEEE 26th International Parallel and Distributed Processing Symposium s. 557 - 568 |
|---|---|
| Hlavní autoři: | , , , , , , |
| Médium: | Konferenční příspěvek Publikace |
| Jazyk: | angličtina |
| Vydáno: |
IEEE
01.05.2012
|
| Témata: | |
| ISBN: | 1467309753, 9781467309752 |
| ISSN: | 1530-2075 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | Clusters of GPUs are emerging as a new computational scenario. Programming them requires the use of hybrid models that increase the complexity of the applications, reducing the productivity of programmers. We present the implementation of OmpSs for clusters of GPUs, which supports asynchrony and heterogeneity for task parallelism. It is based on annotating a serial application with directives that are translated by the compiler. With it, the same program that runs sequentially in a node with a single GPU can run in parallel in multiple GPUs either local (single node) or remote (cluster of GPUs). Besides performing a task-based parallelization, the runtime system moves the data as needed between the different nodes and GPUs minimizing the impact of communication by using affinity scheduling, caching, and by overlapping communication with the computational task. We show several applications programmed with OmpSs and their performance with multiple GPUs in a local node and in remote nodes. The results show good tradeoff between performance and effort from the programmer. |
|---|---|
| AbstractList | Clusters of GPUs are emerging as a new computational scenario. Programming them requires the use of hybrid models that increase the complexity of the applications, reducing the productivity of programmers. We present the implementation of OmpSs for clusters of GPUs, which supports asynchrony and heterogeneity for task parallelism. It is based on annotating a serial application with directives that are translated by the compiler. With it, the same program that runs sequentially in a node with a single GPU can run in parallel in multiple GPUs either local (single node) or remote (cluster of GPUs). Besides performing a task-based parallelization, the runtime system moves the data as needed between the different nodes and GPUs minimizing the impact of communication by using affinity scheduling, caching, and by overlapping communication with the computational task. We show several applications programmed with OmpSs and their performance with multiple GPUs in a local node and in remote nodes. The results show good tradeoff between performance and effort from the programmer. Clusters of GPUs are emerging as a new computational scenario. Programming them requires the use of hybrid models that increase the complexity of the applications, reducing the productivity of programmers. We present the implementation of OmpSs for clusters of GPUs, which supports asynchrony and heterogeneity for task parallelism. It is based on annotating a serial application with directives that are translated by the compiler. With it, the same program that runs sequentially in a node with a single GPU can run in parallel in multiple GPUs either local (single node) or remote (cluster of GPUs). Besides performing a task-based parallelization, the runtime system moves the data as needed between the different nodes and GPUs minimizing the impact of communication by using affinity scheduling, caching, and by overlapping communication with the computational task. We show several applicactions programmed with OmpSs and their performance with multiple GPUs in a local node and in remote nodes. The results show good tradeoff between performance and effort from the programmer. Peer Reviewed |
| Author | Bueno, J. Labarta, J. Planas, J. Martorell, X. Ayguade, E. Duran, A. Badia, R. M. |
| Author_xml | – sequence: 1 givenname: J. surname: Bueno fullname: Bueno, J. email: javier.bueno@bsc.es organization: Barcelona Supercomput. Center, Barcelona, Spain – sequence: 2 givenname: J. surname: Planas fullname: Planas, J. email: judit.planas@bsc.es organization: Barcelona Supercomput. Center, Barcelona, Spain – sequence: 3 givenname: A. surname: Duran fullname: Duran, A. email: alex.duran@bsc.es organization: Barcelona Supercomput. Center, Barcelona, Spain – sequence: 4 givenname: R. M. surname: Badia fullname: Badia, R. M. email: rosa.m.badia@bsc.es organization: Barcelona Supercomput. Center, Artificial Intell. Res. Inst. (IIIA), Barcelona, Spain – sequence: 5 givenname: X. surname: Martorell fullname: Martorell, X. email: xavier.martorell@bsc.es organization: Barcelona Supercomput. Center, Univ. Politec. de Catalunya, Barcelona, Spain – sequence: 6 givenname: E. surname: Ayguade fullname: Ayguade, E. email: eduard.ayguade@bsc.es organization: Barcelona Supercomput. Center, Univ. Politec. de Catalunya, Barcelona, Spain – sequence: 7 givenname: J. surname: Labarta fullname: Labarta, J. email: jesus.labarta@bsc.es organization: Barcelona Supercomput. Center, Univ. Politec. de Catalunya, Barcelona, Spain |
| BookMark | eNpFTstOwkAUHSMmCrJ05aY_ULx33rM0qEhCQhNk3UynU6yhlMwUjX_vGExcnNfinJwxGR36gyfkDmGGCOZhWTwVmxkFpDOhL8jUKA1KGsGlEuaSjDEZBkYJNiI3KBjkFJS4JuMYPwAoMG5uCBahr09uaD99luwu2K5rD7usb7JFsc3m-1McfIjZVzu8Z-vuuIm35Kqx--infzoh25fnt_lrvlovlvPHVe4oqiG3CirqGDiduKoU0rpG4RhPJ7iopOeNqlnjOVikXKZ_TW2s0FhVloNzbELwvOviyZXBOx-cHcretv_hF2mPJqJa6tS5P3da7315DG1nw3cpqVRaaPYDZ7JXvA |
| CODEN | IEEPAD |
| ContentType | Conference Proceeding Publication |
| Contributor | Universitat Politècnica de Catalunya. CAP - Grup de Computació d'Altes Prestacions Universitat Politècnica de Catalunya. Departament d'Arquitectura de Computadors |
| Contributor_xml | – sequence: 1 fullname: Universitat Politècnica de Catalunya. Departament d'Arquitectura de Computadors – sequence: 2 fullname: Universitat Politècnica de Catalunya. CAP - Grup de Computació d'Altes Prestacions |
| Copyright | info:eu-repo/semantics/openAccess |
| Copyright_xml | – notice: info:eu-repo/semantics/openAccess |
| DBID | 6IE 6IL CBEJK RIE RIL XX2 |
| DOI | 10.1109/IPDPS.2012.58 |
| DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume IEEE Xplore All Conference Proceedings IEEE Electronic Library (IEL) IEEE Proceedings Order Plans (POP All) 1998-Present Recercat |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISBN | 9780769546759 0769546757 |
| EndPage | 568 |
| ExternalDocumentID | oai_recercat_cat_2072_202868 6267858 |
| Genre | orig-research |
| GroupedDBID | 29O 6IE 6IF 6IH 6IK 6IL 6IN AAJGR AAWTH ABLEC ADZIZ ALMA_UNASSIGNED_HOLDINGS BEFXN BFFAM BGNUA BKEBE BPEOZ CBEJK CHZPO IEGSK IPLJI OCL RIE RIL ADFMO IERZE XX2 |
| ID | FETCH-LOGICAL-c217t-a70b2c30c82c3bb712dd15c3420745b6e4f7d3fe40a1246153fd9a581bba40cc3 |
| IEDL.DBID | RIE |
| ISBN | 1467309753 9781467309752 |
| ISICitedReferencesCount | 88 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000309131900049&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1530-2075 |
| IngestDate | Fri Nov 07 13:57:18 EST 2025 Wed Aug 27 04:45:00 EDT 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | false |
| IsScholarly | false |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c217t-a70b2c30c82c3bb712dd15c3420745b6e4f7d3fe40a1246153fd9a581bba40cc3 |
| OpenAccessLink | https://recercat.cat/handle/2072/202868 |
| PageCount | 12 |
| ParticipantIDs | ieee_primary_6267858 csuc_recercat_oai_recercat_cat_2072_202868 |
| PublicationCentury | 2000 |
| PublicationDate | 2012-May 2012 |
| PublicationDateYYYYMMDD | 2012-05-01 2012-01-01 |
| PublicationDate_xml | – month: 05 year: 2012 text: 2012-May |
| PublicationDecade | 2010 |
| PublicationTitle | 2012 IEEE 26th International Parallel and Distributed Processing Symposium |
| PublicationTitleAbbrev | ipdps |
| PublicationYear | 2012 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| SSID | ssj0020349 ssj0000781219 |
| Score | 1.851894 |
| Snippet | Clusters of GPUs are emerging as a new computational scenario. Programming them requires the use of hybrid models that increase the complexity of the... Clusters of GPUs are emerging as a new computational scenario. Programming them requires the use of hybrid models that increase the complexity of the... |
| SourceID | csuc ieee |
| SourceType | Open Access Repository Publisher |
| StartPage | 557 |
| SubjectTerms | accelerators Arquitectura de computadors Arquitectures distribuïdes Cluster programming Coherence Computació distribuïda Computational grids (Computer systems) Computer architecture GPGPU computing Graphics processing unit Informàtica Kernel Message systems OpenMP Programming Runtime Àrees temàtiques de la UPC |
| Title | Productive Programming of GPU Clusters with OmpSs |
| URI | https://ieeexplore.ieee.org/document/6267858 https://recercat.cat/handle/2072/202868 |
| WOSCitedRecordID | wos000309131900049&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT8MwDLa2iQMnHhtivJQDJ0S3vpOeBwMuo9KYtFvVpImExB5aW34_dtt1QuLCoVFTNWplu47t2p8B7iXPPC1tYSkRGMtPhbRE5gSWkFxFaCEoY2TVbILPZmK5jOIOPLa1MFrrKvlMj-i0-pefbVRJobIxGt9cBKILXc55XavVxlMItMYlaLLG2SLclRor1UZJ4EFV1BWiPFMl6R7rqZm7B_DN8Vv8FM8p5csdURv4nspL9avzSrXxTE_-98qnMDhU8LG43ZvOoKPX53Cyb-HAmi-6D05cI76izqP7KVVrhQvYxrCXeMEmXyUBKeSMwrXsfbWd5wNYTJ8_Jq9W00XBUuhuFFbKbekqz1YCRym542bIDOX5SBU_kKH2DfLLaN9OHQKXCzyTRWmA5qxMkVfKu4DeerPWl8DCCL0lI9EnMwaXRtLXbii1zBRpDj8dwgORK0FdrXcqLRLCrm4ndOAzXRxcEYoh9IlcybZG1UgaSl39ffkajokndarhDfSKXalv4Uh9F5_57q4ShR9Lt6z3 |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3dT4NADG_mNNGnqZtxfvLgkxEHxwHH83RucU6SzcQ3wh13iYn7yMb8-22BzZj44gMXjnCBtKXXlvZXgBsZZp6WjrCV8I3NUyFtkbm-LWSoIrQQlDGyaDYRjkbi_T2Ka3C3rYXRWhfJZ_qeTot_-dlcrSlU1kHjOxS-2IFdn3PmltVa24gKwdYwAier3C1CXinRUh2UhdAvyroClGiqJd2gPVVz9gO_2RnED_GYkr7YPTWCr6vVWv3qvVJsPb3G_176EFo_NXxWvN2djqCmZ8fQ2DRxsKpvugluXGK-otaj-ylZa4oLrLmxnuI3q_u5JiiFlUUBW-t1uhivWvDWe5x0-3bVR8FW6HDkdho6kinPUQJHKUOXZcgO5XGkCvdloLlBjhnNndQleDnfM1mU-mjQyhS5pbwTqM_mM30KVhChv2QkemXG4NJIcs0CqWWmSHfwtA23RK4EtbVeqjRPCL16O6EDn8lwYCIQbWgSuZJFiauRVJQ6-_vyNez3Jy_DZDgYPZ_DAfGnTDy8gHq-XOtL2FNf-cdqeVWIxTeBdbA- |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=2012+IEEE+26th+International+Parallel+and+Distributed+Processing+Symposium&rft.atitle=Productive+Programming+of+GPU+Clusters+with+OmpSs&rft.au=Bueno%2C+J.&rft.au=Planas%2C+J.&rft.au=Duran%2C+A.&rft.au=Badia%2C+R.+M.&rft.date=2012-05-01&rft.pub=IEEE&rft.isbn=9781467309752&rft.issn=1530-2075&rft.spage=557&rft.epage=568&rft_id=info:doi/10.1109%2FIPDPS.2012.58&rft.externalDocID=6267858 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1530-2075&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1530-2075&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1530-2075&client=summon |