Publication:
An autotuning approach to select the inter-GPU communication library on heterogeneous systems

Simple item page
dc.contributor.authorCámara, Jesús
dc.contributor.authorCuenca Muñoz, Antonio Javier
dc.contributor.authorCuenca, Javier
dc.contributor.authorBoratto, Murilo
dc.contributor.authorVicente Jaén, Arturo
dc.contributor.authorGalindo Garre, Víctor
dc.contributor.departmentIngeniería y Tecnología de Computadores
dc.date.accessioned2025-11-11T11:51:40Z
dc.date.available2025-11-11T11:51:40Z
dc.date.copyright© The Author(s) 2024
dc.date.issued2024-12-12
dc.description.abstractIn this work, an automatic optimisation approach for parallel routines on multi-GPU systems is presented. Several inter-GPU communication libraries (such as CUDA- Aware MPI or NCCL) are used with a set of routines to perform the numerical oper- ations among the GPUs located on the compute nodes. The main objective is the selection of the most appropriate communication library, the number of GPUs to be used and the workload to be distributed among them in order to reduce the cost of data movements, which represent a large percentage of the total execution time. To this end, a hierarchical modelling of the execution time of each routine to be opti- mised is proposed, combining experimental and theoretical approaches. The results show that near-optimal decisions are taken in all the scenarios analysed.
dc.formatapplication/pdf
dc.format.extent16
dc.identifier.citationJournal of Supercomputing, 2025, Vol. 81, 283
dc.identifier.doihttps://doi.org/10.1007/s11227-024-06794-3
dc.identifier.eissn1573-0484
dc.identifier.issn0920-8542
dc.identifier.urihttp://hdl.handle.net/10201/172289
dc.languageeng
dc.publisherSpringer
dc.relationEste trabajo cuenta con el apoyo de la subvención PID2022-136315OB-I00 y la subvención PID2022-142292NB-I00, ambas financiadas por el Ministerio de Ciencia, Innovación y Universidades (MCIN/AEI/10.13039/501100011033/) y por «FEDER Una forma de hacer Europa», UE.
dc.relation.publisherversionhttps://link.springer.com/article/10.1007/s11227-024-06794-3
dc.rightsAttribution 4.0 International*
dc.rights.accessRightsinfo:eu-repo/semantics/openAccess
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/*
dc.subjectAutotuning
dc.subjectCommunication libraries
dc.subjectMulti GPU
dc.subjectHeterogeneous computing
dc.subject.odsObjetivo 9: Infraestructura
dc.titleAn autotuning approach to select the inter-GPU communication library on heterogeneous systems
dc.typeinfo:eu-repo/semantics/article
dspace.entity.typePublicationes
relation.isAuthorOfPublication7f8123b6-544c-4956-8229-538e4d177c31
relation.isAuthorOfPublication89b3f3c2-f773-43d1-bb21-8035df444cdc
relation.isAuthorOfPublicationef5d5478-2d76-4064-a920-b037ac19f5fe
relation.isAuthorOfPublication65345b4c-0fae-4b76-8f83-8b78e5b08157
relation.isAuthorOfPublication.latestForDiscovery7f8123b6-544c-4956-8229-538e4d177c31
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2025_revista_Q2_Journal_of_Supercomputing.pdf
Size:
1.22 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.37 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Artículos