A framework for direct and transparent data exchange of filter-stream applications in multi-GPUs architectures

Página do item simplificado
dc.creatorGabriel Ramos
dc.creatorGuilherme Neri Andrade
dc.creatorRafael Sachetto
dc.creatorDaniel Madeira
dc.creatorRenan Carvalho
dc.creatorRenato Antonio Celso Ferreira
dc.creatorFernando Mourão
dc.creatorLeonardo Rocha
dc.date.accessioned2025-03-06T21:16:02Z
dc.date.accessioned2025-09-08T23:39:23Z
dc.date.available2025-03-06T21:16:02Z
dc.date.issued2017-06-09
dc.description.sponsorshipCNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico
dc.description.sponsorshipFAPEMIG - Fundação de Amparo à Pesquisa do Estado de Minas Gerais
dc.description.sponsorshipCAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
dc.format.mimetypepdf
dc.identifier.doihttps://doi.org/10.1016/j.procs.2017.05.144
dc.identifier.urihttps://hdl.handle.net/1843/80548
dc.languageeng
dc.publisherUniversidade Federal de Minas Gerais
dc.relationPrograma Institucional de Internacionalização – CAPES - PrInt
dc.relation.ispartofInternational Conference on Computational Science
dc.rightsAcesso Aberto
dc.subjectComputação
dc.subjectCluster (Sistema de computador)
dc.subjectCiência da computação
dc.subject.otherMultiple GPUs
dc.subject.otherFilter Stream Application
dc.subject.otherHeterogeneous Architectures
dc.titleA framework for direct and transparent data exchange of filter-stream applications in multi-GPUs architectures
dc.typeArtigo de evento
local.citation.epage1651
local.citation.spage1642
local.description.resumoThe massive data generation has been pushing for significant advances in computing architectures, reflecting in heterogeneous architectures composed by different types of processing units. The filter-stream paradigm is typically used to exploit the parallel processing power of these new architectures. The efficiency of applications in this paradigm is achieved by exploring a set of interconnected computers (cluster) using filters and communication between them in a coordinated way. In this work we propose, implement and test a generic abstraction for direct and transparent data exchange of filter-stream applications in heterogeneous cluster with multi-GPU (Graphics Processing Units). This abstraction allows hiding from the programmers all the low-level implementation details related to GPU communication and the control related to the location of filters. Further, we consolidate such abstraction into a framework. Empirical assessments using a real application show that the proposed abstraction layer eases the implementation of filter-stream applications without compromising the overall application performance.
local.identifier.orcidhttps://orcid.org/0000-0003-0800-5984
local.identifier.orcidhttps://orcid.org/0000-0002-4372-8996
local.publisher.countryBrasil
local.publisher.departmentICX - DEPARTAMENTO DE CIÊNCIA DA COMPUTAÇÃO
local.publisher.initialsUFMG
local.url.externahttps://www.sciencedirect.com/science/article/pii/S1877050917307160

Arquivos

Pacote original

Agora exibindo 1 - 1 de 1
Imagem de Miniatura
Nome:
A Framework for Direct and Transparent Data Exchange of Filter-stream Applications in Multi-GPUs Architectures (1).pdf
Tamanho:
691.88 KB
Formato:
Adobe Portable Document Format
Baixar

Licença do pacote

Agora exibindo 1 - 1 de 1
Imagem de Miniatura
Nome:
License.txt
Tamanho:
1.99 KB
Formato:
Plain Text
Descrição:
Baixar

Coleções

Artigo de Evento

Logo UFMG