Close

1. Identity statement
Reference TypeJournal Article
Siteplutao.sid.inpe.br
Holder Codeisadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S
IdentifierJ8LNKAN8RW/39RQF8Q
Repositorydpi.inpe.br/plutao/2011/06.11.03.44.28   (restricted access)
Last Update2012:02.03.12.57.05 (UTC) administrator
Metadata Repositorydpi.inpe.br/plutao/2011/06.11.03.44.29
Metadata Last Update2021:01.02.03.52.36 (UTC) administrator
Secondary KeyINPE--PRE/
DOI10.1504/ijhpsa.2011.040462
ISSN1751-6528
Labellattes: 3723124745578916 3 FazendaPaKaRoMoNa:2011:ChSoIm
Citation KeyFazendaPaKaRoMoNa:2011:ChSoIm
TitleChallenges and solutions to improve the scalability of an operational regional meteorological forecasting model
Year2011
Access Date2024, May 02
Secondary TypePRE PI
Number of Files1
Size7081 KiB
2. Context
Author1 Fazenda, Alvaro Luiz
2 Panetta, Jairo
3 Katsurayama, Daniel Massaru
4 Rodriguez, Luiz Flavio
5 Motta, Luis Filipe
6 Navaux, Philippe Olivier Alexandre
Group1
2 DMD-CPT-INPE-MCT-BR
3 DMD-CPT-INPE-MCT-BR
4 DMD-CPT-INPE-MCT-BR
Affiliation1
2 Instituto Nacional de Pesquisas Espaciais (INPE)
3 Instituto Nacional de Pesquisas Espaciais (INPE)
4 Instituto Nacional de Pesquisas Espaciais (INPE)
Author e-Mail Address1
2
3 daniel.massaru@cptec.inpe.br
e-Mail Addressdaniel.massaru@cptec.inpe.br
JournalInternational Journal of High Performance Systems Architecture
Volume3
Number2/3
Pages87-97
Secondary MarkB1_CIÊNCIA_DA_COMPUTAÇÃO B3_ENGENHARIAS_IV
History (UTC)2011-06-11 17:43:46 :: lattes -> marciana :: 2011
2011-06-20 18:19:39 :: marciana -> valdirene :: 2011
2011-11-03 16:34:47 :: valdirene -> banon :: 2011
2011-11-18 16:53:20 :: banon -> valdirene :: 2011
2012-02-03 12:57:17 :: valdirene -> administrator :: 2011
2021-01-02 03:52:36 :: administrator -> valdirene :: 2011
3. Content and structure
Is the master or a copy?is the master
Content Stagecompleted
Transferable1
Content TypeExternal Contribution
Version Typepublisher
Keywordsparallel scalability
regional weather forecasting models
AbstractThis work investigates the parallel scalability of BRAMS, a limited area weather forecasting production code, from O(100) cores to O(1,000) cores on large grids (20 km and 10 km resolution runs over South America). Initial experiments show lack of scalability at modest core count. Execution time profiling and source code examination revealed the causes of the limited scalability: sequential algorithms and extensive memory requirements at scarcely used phases of the computation. As processor count increases, these 'secondary' phases dominate execution time. Algorithm replacement and memory reduction generate a new code version that possesses strong and weak scaling. The new version achieved a speed-up of 6 from 100 to 700 processors on the 20 km resolution grid and a speed-up of 6.9 on the same processor range on the 10 km resolution grid. Results were confirmed at another machine with a distinct architecture. Further experiments show that the scalability of the 20 km resolution case is limited by load unbalancing at the most demanding computational phase. Copyright © 2011 Inderscience Enterprises Ltd.
AreaMET
Arrangementurlib.net > BDMCI > Fonds > Produção anterior à 2021 > DIDMD > Challenges and solutions...
doc Directory Contentaccess
source Directory Contentthere are no files
agreement Directory Contentthere are no files
4. Conditions of access and use
Languageen
User Groupadministrator
lattes
marciana
banon
valdirene
Visibilityshown
Archiving Policydenypublisher denyfinaldraft6
Read Permissiondeny from all and allow from 150.163
Update Permissionnot transferred
5. Allied materials
Next Higher Units8JMKD3MGPCW/43SKC35
URL (untrusted data)http://www.inderscience.com/browse/index.php?journalID=213
DisseminationCOMPENDEX
Host Collectiondpi.inpe.br/plutao@80/2008/08.19.15.01
6. Notes
NotesSetores de Atividade: Pesquisa e desenvolvimento científico.
Empty Fieldsalternatejournal archivist callnumber copyholder copyright creatorhistory descriptionlevel format isbn lineage mark mirrorrepository month nextedition orcid parameterlist parentrepositories previousedition previouslowerunit progress project readergroup resumeid rightsholder schedulinginformation secondarydate session shorttitle sponsor subject targetfile tertiarymark tertiarytype typeofwork
7. Description control
e-Mail (login)valdirene
update 


Close