KISTI repository
BROWSE
download0
view691
This item is licensed Korea Open Government License
- dc.contributor.author
- 차광호
- dc.contributor.author
- 홍정우
- dc.contributor.author
- 홍태영
- dc.date.accessioned
- 2019-08-28T07:40:24Z
- dc.date.available
- 2019-08-28T07:40:24Z
- dc.date.issued
- 2004-12-08
- dc.identifier.issn
- 0302-9743
- dc.identifier.uri
- https://repository.kisti.re.kr/handle/10580/13578
- dc.identifier.uri
- http://www.ndsl.kr/ndsl/search/detail/article/articleSearchResultDetail.do?cn=NART23551648
- dc.description.abstract
- Because many scientific applications require large data processing, the importance of parallel I/O has been increasingly recognized. Collective I/O is one of the considerable features of parallel I/O. In this paper we suggest the subgroup method for collective I/O. It is the way of using collective I/Os of MPI effectively in terms of application programs. From the experimental results, we can conclude that the subgroup method for collective I/O is more efficient than plain collective I/O and more reliable than non-collective I/O.
- dc.language
- eng
- dc.relation.ispartofseries
- Lecture Notes of Computer Science(LNCS)
- dc.title
- The Subgroup Method for Collective I/O
- dc.citation.endPage
- 304
- dc.citation.number
- 1
- dc.citation.startPage
- 301
- dc.citation.volume
- 3320
- dc.subject.keyword
- Collective I/O
- dc.subject.keyword
- Cluster computing
- dc.subject.keyword
- MPI
- Appears in Collections:
- 7. KISTI 연구성과 > 학술지 발표논문
- Files in This Item:
- There are no files associated with this item.
