- -

Analyzing the performance/power tradeoff of the rCUDA middleware for future exascale systems

RiuNet: Institutional repository of the Polithecnic University of Valencia

Share/Send to

Cited by

Statistics

Analyzing the performance/power tradeoff of the rCUDA middleware for future exascale systems

Show simple item record

Files in this item

dc.contributor.author Reaño González, Carlos es_ES
dc.contributor.author Prades, Javier es_ES
dc.contributor.author Silla Jiménez, Federico es_ES
dc.date.accessioned 2021-01-14T04:33:05Z
dc.date.available 2021-01-14T04:33:05Z
dc.date.issued 2019-10 es_ES
dc.identifier.issn 0743-7315 es_ES
dc.identifier.uri http://hdl.handle.net/10251/158950
dc.description.abstract [EN] The computing power of supercomputers and data centers has noticeably grown during the last decades at the cost of an ever increasing energy demand. The need for energy (and power) of these facilities has finally limited the evolution of high performance computing, making that many researchers are concerned not only about performance but also about energy efficiency. However, despite the many concerns about energy consumption, the search for computing power continues. In this regard, the research on exascale systems, able to deliver 10(18) floating point operations per second, has reached a widely consensus that these systems should operate within a maximum power budget of 20 megawatts. Many efficiency improvements are necessary for achieving this goal. One of these improvements is the usage of ARM low-power processors, as the Mont-Blanc project proposes. In this paper we analyze the combined use of ARM processors with the rCUDA remote GPU virtualization middleware as a way to improve efficiency even more. Results show that it is possible to speed up applications by almost 8x while reducing energy consumption up to 35% when rCUDA is used to access high-end GPUs. These improvements are achieved while maintaining a feasible average power consumption level for future exascale systems. es_ES
dc.description.sponsorship This work was funded by the Generalitat Valenciana under Grant PROMETEO/2017/077. Authors are also grateful for the generous support provided by Mellanox Technologies Inc and for the equipment donated by NVIDIA Corporation. es_ES
dc.language Inglés es_ES
dc.publisher Elsevier es_ES
dc.relation GENERALITAT VALENCIANA/PROMETEO/2017/077 es_ES
dc.relation.ispartof Journal of Parallel and Distributed Computing es_ES
dc.rights Reconocimiento - No comercial - Sin obra derivada (by-nc-nd) es_ES
dc.subject GPU virtualization es_ES
dc.subject HPC es_ES
dc.subject Energy es_ES
dc.subject Exascale es_ES
dc.subject.classification ARQUITECTURA Y TECNOLOGIA DE COMPUTADORES es_ES
dc.title Analyzing the performance/power tradeoff of the rCUDA middleware for future exascale systems es_ES
dc.type Artículo es_ES
dc.identifier.doi 10.1016/j.jpdc.2019.04.021 es_ES
dc.rights.accessRights Embargado es_ES
dc.date.embargoEndDate 2021-10-31 es_ES
dc.contributor.affiliation Universitat Politècnica de València. Departamento de Informática de Sistemas y Computadores - Departament d'Informàtica de Sistemes i Computadors es_ES
dc.description.bibliographicCitation Reaño González, C.; Prades, J.; Silla Jiménez, F. (2019). Analyzing the performance/power tradeoff of the rCUDA middleware for future exascale systems. Journal of Parallel and Distributed Computing. 132:344-362. https://doi.org/10.1016/j.jpdc.2019.04.021 es_ES
dc.description.accrualMethod S es_ES
dc.relation.publisherversion https://doi.org/10.1016/j.jpdc.2019.04.021 es_ES
dc.description.upvformatpinicio 344 es_ES
dc.description.upvformatpfin 362 es_ES
dc.type.version info:eu-repo/semantics/publishedVersion es_ES
dc.description.volume 132 es_ES
dc.relation.pasarela S\399118 es_ES
dc.contributor.funder Generalitat Valenciana es_ES


This item appears in the following Collection(s)

Show simple item record