Mostrar el registro sencillo del ítem
A scheduling theory framework for GPU tasks efficient execution
dc.contributor.author | Lázaro Muñoz, Antonio José | |
dc.contributor.author | López Albelda, Bernabé | |
dc.contributor.author | González-Linares, José María | |
dc.contributor.author | Guil-Mata, Nicolás | |
dc.date.accessioned | 2018-07-16T11:34:20Z | |
dc.date.available | 2018-07-16T11:34:20Z | |
dc.date.created | 2018-07-13 | |
dc.date.issued | 2018-07-16 | |
dc.identifier.uri | https://hdl.handle.net/10630/16279 | |
dc.description.abstract | Concurrent execution of tasks in GPUs can reduce the computation time of a workload by overlapping data transfer and execution commands. However it is difficult to implement an efficient run- time scheduler that minimizes the workload makespan as many execution orderings should be evaluated. In this paper, we employ scheduling theory to build a model that takes into account the device capabili- ties, workload characteristics, constraints and objec- tive functions. In our model, GPU tasks schedul- ing is reformulated as a flow shop scheduling prob- lem, which allow us to apply and compare well known methods already developed in the operations research field. In addition we develop a new heuristic, specif- ically focused on executing GPU commands, that achieves better scheduling results than previous tech- niques. Finally, a comprehensive evaluation, showing the suitability and robustness of this new approach, is conducted in three different NVIDIA architectures (Kepler, Maxwell and Pascal). | en_US |
dc.description.sponsorship | Proyecto TIN2016- 0920R, Universidad de Málaga (Campus de Excelencia Internacional Andalucía Tech) y programa de donación de NVIDIA Corporation. | en_US |
dc.language.iso | eng | en_US |
dc.rights | info:eu-repo/semantics/openAccess | en_US |
dc.rights.uri | http://creativecommons.org/licenses/by-nc-nd/4.0/ | * |
dc.subject | Arquitectura de redes informáticas - Congresos | en_US |
dc.subject.other | GPU | en_US |
dc.subject.other | Task reordering | en_US |
dc.title | A scheduling theory framework for GPU tasks efficient execution | en_US |
dc.type | info:eu-repo/semantics/conferenceObject | en_US |
dc.centro | E.T.S.I. Informática | en_US |
dc.relation.eventtitle | Jornadas Sarteco | en_US |
dc.relation.eventplace | Teruel | en_US |
dc.relation.eventdate | 12-14 de Septiembre de 2018 | en_US |
dc.rights.cc | Attribution-NonCommercial-NoDerivatives 4.0 Internacional | * |