Close
Metadata

1. Identity statement
Reference TypeJournal Article
Sitemtc-m16d.sid.inpe.br
Holder Codeisadg {BR SPINPE} ibi 8JMKD3MGPCW/3DT298S
Identifier8JMKD3MGP7W/3DETRSE
Repositorysid.inpe.br/mtc-m19/2013/01.28.13.41   (restricted access)
Last Update2013:02.06.13.15.04 (UTC) administrator
Metadata Repositorysid.inpe.br/mtc-m19/2013/01.28.13.41.26
Metadata Last Update2018:06.05.04.13.55 (UTC) administrator
Secondary KeyINPE--PRE/
DOI10.1016/j.scico.2012.07.005
ISSN0167-6423
Citation KeyRodriguesNavaPaneMend:2013:PrOrMP
TitlePreserving the original MPI semantics in a virtualized processor environment
Year2013
MonthApr.
Access Date2024, Apr. 18
Secondary TypePRE PI
Number of Files1
Size580 KiB
2. Context
Author1 Rodrigues, Eduardo R.
2 Navaux, Philippe O. A.
3 Panetta, Jairo
4 Mendes, Celso L.
Group1
2
3 CPT-CPT-INPE-MCTI-GOV-BR
Affiliation1 IBM Research, Brazil
2 Institute of Informatics, UFRGS, Brazil
3 Instituto Nacional de Pesquisas Espaciais (INPE)
4 Parallel Programming Laboratory, UIUC, USA
Author e-Mail Address1 edrodri@br.ibm.com
JournalScience of Computer Programming
Volume78
Number4
Pages412–421
Secondary MarkB1_CIÊNCIA_DA_COMPUTAÇÃO B1_ENGENHARIAS_IV A1_INTERDISCIPLINAR
History (UTC)2013-02-06 13:12:35 :: marciana :: 2012 -> 2013
2013-02-06 13:15:04 :: marciana -> administrator :: 2013
2018-06-05 04:13:55 :: administrator -> marcelo.pazos@inpe.br :: 2013
3. Content and structure
Is the master or a copy?is the master
Content Stagecompleted
Transferable1
Content TypeExternal Contribution
Version Typepublisher
Keywordsprocessor virtualization
MPI
user-level threads
privatization
TLS
AbstractProcessor virtualization is a technique in which a programmer divides a computation into many entities, which are mapped to the available processors. The number of these entities, referred to as virtual processors, is typically larger than the number of physical processors. For an MPI program, the user decomposes the computation into more MPI tasks than physical processors. This approach allows overlapping computation and communication, and enables load balancing. User-level threads are often used to implement these virtual processors because they are generally faster to create, manage and migrate than heavy processes or kernel threads. However, these threads present issues concerning private data because they break the private address space assumption typically made by MPI programs. In this paper, we propose a new approach to privatize data in user-level threads. This approach is based on thread-local storage (TLS), which is often used by kernel threads. We apply this technique so that MPI programs can be executed in a virtualized environment while preserving their original semantics. We show that this alternative has a more efficient context switch and lower migration cost and is simpler to implement than other approaches.
AreaMET
Arrangementurlib.net > BDMCI > Fonds > Produção anterior à 2021 > CGCPT > Preserving the original...
doc Directory Contentaccess
source Directory Contentthere are no files
agreement Directory Content
agreement.html 28/01/2013 11:41 1.0 KiB 
4. Conditions of access and use
Languageen
Target FileRodrigues.pdf
User Groupadministrator
marciana
Visibilityshown
Archiving Policydenypublisher denyfinaldraft24
Read Permissiondeny from all and allow from 150.163
Update Permissionnot transferred
5. Allied materials
Mirror Repositoryiconet.com.br/banon/2006/11.26.21.31
Next Higher Units8JMKD3MGPCW/3EUPEJL
DisseminationWEBSCI; PORTALCAPES.
Host Collectionsid.inpe.br/mtc-m19@80/2009/08.21.17.02
6. Notes
Empty Fieldsalternatejournal archivist callnumber copyholder copyright creatorhistory descriptionlevel e-mailaddress format isbn label lineage mark nextedition notes orcid parameterlist parentrepositories previousedition previouslowerunit progress project readergroup resumeid rightsholder schedulinginformation secondarydate session shorttitle sponsor subject tertiarymark tertiarytype typeofwork url
7. Description control
e-Mail (login)marcelo.pazos@inpe.br
update 


Close