Fechar

1. Identificação
Tipo de ReferênciaArtigo em Evento (Conference Proceedings)
Sitemtc-m12.sid.inpe.br
Código do Detentorisadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S
Identificador6qtX3pFwXQZ3r59YD6/GNUE5
Repositóriosid.inpe.br/iris@1912/2005/07.19.21.30.47   (acesso restrito)
Última Atualização2016:02.02.13.28.17 (UTC) marciana
Repositório de Metadadossid.inpe.br/iris@1912/2005/07.19.21.30.55
Última Atualização dos Metadados2018:06.05.00.52.17 (UTC) administrator
Chave SecundáriaINPE-7948-PRE/3783
Rótulo8886
Chave de CitaçãoMendesReed:1997:AuScPr
TítuloAutomated scalability prediction via data parallel compiler support
Ano1997
Data Secundária20000906
Data de Acesso11 maio 2024
Tipo SecundárioPRE CN
Número de Arquivos1
Tamanho567 KiB
2. Contextualização
Autor1 Mendes, Celso Luiz
2 Reed, Daniel A
Grupo1 DPI-INPE-MCT-BR
Nome do EventoSimpósio Brasileiro de Arquitetura de Computadores: Processamento de Alto Desempenho.
Localização do EventoCampos do Jordao
Data7-10 out. 1997
Páginas349-364
Título do LivroAnais
Histórico (UTC)2005-07-19 21:30:55 :: administrator -> jefferson ::
2006-05-30 12:23:15 :: jefferson -> administrator ::
2008-06-09 21:43:39 :: administrator -> jefferson ::
2010-07-07 18:49:15 :: jefferson -> administrator ::
2015-05-04 14:17:06 :: administrator -> marciana :: 1997
2016-02-02 13:28:17 :: marciana -> administrator :: 1997
2018-06-05 00:52:17 :: administrator -> jefferson :: 1997
3. Conteúdo e estrutura
É a matriz ou uma cópia?é a matriz
Estágio do Conteúdoconcluido
Transferível1
Tipo do ConteúdoExternal Contribution
Palavras-ChavePROCESSAMENTO DIGITAL DE IMAGENS
PROCESSAMENTO PARALELO
LINGUAGEM COMPUTACIONAL (COMPUTACAO)
COMPILADORES
DATA PROCESSING
PARALLEL PROCESSING
PROGRAMMING LANGUAGEM
COMPILERS
COMPUTER SYSTEMS PERFORMANCE
ResumoDespite the performance potential of multicomputers, several factors have limited their widespread adoption. Of these, performance variability is among the most significant Execution of some programs may yield only a small fraction of peak system performance, whereas others approach the system's theoretical performance peak. Moreover, the observed performance may change substantially as application program parameters vary. Data parallel languages, which facilitate the programming of multicomputers, increase the semantic distance between the program's source code and its observable performance, thus aggravating the performance problem. In this paper, we propose a new methodology to automaticaly predict the performance scalability of data parallel applications on multicomputers. Our technique represents the execution time of program as a symbolic expression that is a function of the number of processors (P), problem size (N), and other system-dependent parameters. This methodology is based on information collected at compile time. By extending an existing data parallel compiler (Fortran D95), we derive, during compilation, a symbolic: model that represents the cost of each high-level program section and, inductively, of the complete program. These symbolic expression may be simplified externally with current symbolic tools. Predicting performance of the program for a given pair (P, N)requires simply the evaluation of its corresponding cost expression. We validate our implementation by predicting scalability of a variety of loop nests, with distinct computation and communication patterns.
ÁreaSRE
Arranjourlib.net > BDMCI > Fonds > Produção anterior à 2021 > DIDPI > Automated scalability prediction...
Conteúdo da Pasta docacessar
Conteúdo da Pasta sourcenão têm arquivos
Conteúdo da Pasta agreementnão têm arquivos
4. Condições de acesso e uso
Idiomaen
Arquivo AlvoINPE 7948.pdf
Grupo de Usuáriosadministrator
jefferson
Grupo de Leitoresadministrator
jefferson
marciana
Visibilidadeshown
Detentor da CópiaSID/SCD
Permissão de Leituradeny from all and allow from 150.163
Permissão de Atualizaçãonão transferida
5. Fontes relacionadas
Unidades Imediatamente Superiores8JMKD3MGPCW/3EQCCU5
Acervo Hospedeirosid.inpe.br/banon/2001/04.06.10.52
6. Notas
Campos Vaziosaffiliation archivingpolicy archivist callnumber copyright creatorhistory descriptionlevel dissemination doi e-mailaddress edition editor electronicmailaddress format isbn issn lineage mark mirrorrepository nextedition notes numberofvolumes orcid organization parameterlist parentrepositories previousedition previouslowerunit progress project publisher publisheraddress resumeid rightsholder schedulinginformation secondarymark serieseditor session shorttitle sponsor subject tertiarymark tertiarytype type url versiontype volume
7. Controle da descrição
e-Mail (login)jefferson
atualizar 


Fechar