| |
|
|
|
Registro Completo |
|
Biblioteca(s): |
Embrapa Agricultura Digital. |
|
Data corrente: |
03/07/1996 |
|
Data da última atualização: |
16/08/2007 |
|
Autoria: |
MENDES, C. L. |
|
Título: |
Integrating message-passing with vector architectures. |
|
Ano de publicação: |
1995 |
|
Fonte/Imprenta: |
In: SIMPÓSIO BRASILEIRO DE ARQUITETURA DE COMPUTADORES - PROCESSAMENTO DE ALTO DESEMPENHO, 7.; CONGRESSO BRASILEIRO DA SOCIEDADE BRASILEIRA DE COMPUTAÇÃO, 15., 1995, Canela. Anais... Porto Alegre: UFRGS, Instituto de Informática,1995. |
|
Páginas: |
p.151-165. |
|
Idioma: |
Inglês |
|
Conteúdo: |
Vector architectures provide excellent computational throughput, while sucessfully tolerating memory latency by pipelining memory accesses. In this paper, we propose a generalization of vector architectures to message-passing multicomputers, which combines the efficiency of vector computation whith the scalability of distributed-memory systems. In our proposed architecture, each node is a conventional vector processsor (with chaining capability and pipelined functional units) augmented by native instructions to send and receive messages through vector registers. In this scheme, inter-node communication can be performed via vector-send/receive instructions, gaining the benefits of communication pipelining, reduced memory copies (memory-to-register-to-register instead of memory-to-memory-to-cache), and lower communication latency (due to tight processor-communication coupling). We show that this strong integration between functional and communication units can lead to substantial performance improvement over conventional message-passing multicomputers. We model pipelined computation-communication systems both analyticaly and with a detailed instruction-level simulation, and compare this simulation data with empirical results from an intel paragon. Preliminary data from a matrix multiplication example indicates our proposed vector-parallel architecture offers significant scalability benefits over existing message-passing systems. |
|
Categoria do assunto: |
-- |
|
Marc: |
LEADER 01994naa a2200133 a 4500
001 1002839
005 2007-08-16
008 1995 bl uuuu u00u1 u #d
100 1 $aMENDES, C. L.
245 $aIntegrating message-passing with vector architectures.
260 $c1995
300 $ap.151-165.
520 $aVector architectures provide excellent computational throughput, while sucessfully tolerating memory latency by pipelining memory accesses. In this paper, we propose a generalization of vector architectures to message-passing multicomputers, which combines the efficiency of vector computation whith the scalability of distributed-memory systems. In our proposed architecture, each node is a conventional vector processsor (with chaining capability and pipelined functional units) augmented by native instructions to send and receive messages through vector registers. In this scheme, inter-node communication can be performed via vector-send/receive instructions, gaining the benefits of communication pipelining, reduced memory copies (memory-to-register-to-register instead of memory-to-memory-to-cache), and lower communication latency (due to tight processor-communication coupling). We show that this strong integration between functional and communication units can lead to substantial performance improvement over conventional message-passing multicomputers. We model pipelined computation-communication systems both analyticaly and with a detailed instruction-level simulation, and compare this simulation data with empirical results from an intel paragon. Preliminary data from a matrix multiplication example indicates our proposed vector-parallel architecture offers significant scalability benefits over existing message-passing systems.
773 $tIn: SIMPÓSIO BRASILEIRO DE ARQUITETURA DE COMPUTADORES - PROCESSAMENTO DE ALTO DESEMPENHO, 7.; CONGRESSO BRASILEIRO DA SOCIEDADE BRASILEIRA DE COMPUTAÇÃO, 15., 1995, Canela. Anais... Porto Alegre: UFRGS, Instituto de Informática,1995.
Download
Esconder MarcMostrar Marc Completo |
|
Registro original: |
Embrapa Agricultura Digital (CNPTIA) |
|
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
URL |
Voltar
|
|
|
| Registros recuperados : 2 | |
| 1. |  | BENKO-ISEPPON, A. M.; KIDO, E. A.; PANDOLFI, V.; BARBOSA, P. K. A; BELARMINO, L. C.; MONTE, S. J. H. do; BRANDÃO, R. M. S. de; ARAUJO, A. de S.; CASTRO, J. A. F. de; SOARES-CAVALCANTI, N. da M.; SILVA, A. R. da; CALSA JUNIOR, T.; ROCHA, M. de M.; WINTER, P.; KAHL, G.; ROTTER, B.; HORRES, R.; MOLINA, C.; JUNGMANN, R.; AMORIM, L. L. B.; ONOFRE, A. V. C.; FERREIRA-NETO, J. R. C.; GRANJEIRO, T. B.; LIMA, A. S.; LOBO, M. D. P.; HOULLOU-KIDO, L. M.; CARVALHO, R. de; WANDERLEY-NOGUEIRA, A. C.; BARROS, P. dos S.; VIEIRA-MELLO, G. S.; BRASILEIRO-VIDAL, A. C.; BORTOLETI, K. C. de A.; PEDROSA-HARAND, A.; ANDRADE, P. P. de; ANDRADE, G. P. de; PIO-RIBEIRO, G.; SITTOLIN, I. M.; FREIRE FILHO, F. R. Brazilian cowpea transcriptome project: over 20 million expressed sequence tags to understand salinity and virus resistance. In: CONGRESSO BRASILEIRO DE BIOTECNOLOGIA, 3., 2010. Fortaleza. Programa e resumos. Brasília, DF: SBBiotec, 2010. p. 97-98.| Tipo: Resumo em Anais de Congresso |
| Biblioteca(s): Embrapa Meio-Norte. |
|    |
| 2. | | KIDO, E. A.; PANDOLFI, V.; BARBOSA, P. K. A.; SILVA, L. C. B. da; SILVA, A. B. da; MONTE, S. J. H. do; BRANDÃO, R. M. S. de S.; ARAUJO, A. de S.; CASTRO, J. A. F. de; SOARES-CAVALCANTI, N. da M.; CALSA-JUNIOR, T.; ROCHA, M. M.; WINTER, P.; KAHL, G.; ROTTER, B.; HORRES, R.; MOLINA, C.; JUNGMANN, R.; AMORIM, L. L. B.; ONOFRE, A. V. C.; FARIAS NETO, J. C.; GRANJEIRO, T. B.; LIMA, A. S.; LOBO, M. D. P.; HOULLOU-KIDO, L. M.; CARVALHO, R. de; WANDERLEY-NOGUEIRA, A. C.; BARROS, P. dos S.; VIEIRA-MELLO, G. S.; BRASILEIRO-VIDAL, A. C.; BORTOLETI, K. C. de A.; PEDROSA-HARAND, A.; ANDRADE, P. P. de; ANDRADE, G. P. de; PIO-RIBEIRO, G.; SITTOLIN, I. M.; FREIRE FILHO, F. R.; CASTRO, L. A. B. de; BENKO-ISEPPON, A. M. Brazilian cowpea transcriptome project: over five million expressed sequence tags to understand salinity and virus resistance. In: CONGRESSO BRASILEIRO DE FISIOLOGIA VEGETAL, 12., 2009, Fortaleza. Desafios para a produção de alimentos e bioenergia: livro de resumos. Fortaleza: SBFV: UFC: Embrapa Agroindústria Tropical, 2009. p. 25-26. Resumo 61.| Tipo: Resumo em Anais de Congresso |
| Biblioteca(s): Embrapa Meio-Norte. |
| |
| Registros recuperados : 2 | |
|
| Nenhum registro encontrado para a expressão de busca informada. |
|
|
