Algorithmic redistribution methods for block-cyclic decompositions

This article presents various data redistribution methods for block-partitioned linear algebra algorithms operating on dense matrices that are distributed in a block-cyclic fashion. Because the algorithmic partitioning unit and the distribution blacking factor are most often chosen to be equal, seve...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:IEEE transactions on parallel and distributed systems Ročník 10; číslo 12; s. 1201 - 1216
Hlavní autori: Petitet, A.P., Dongarra, J.J.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: IEEE 01.12.1999
Predmet:
Algorithm design and analysis
Algorithms
Alignment
Architecture
Blocking
Concurrent computing
Constrictions
Distributed computing
Linear algebra
Mathematical analysis
Matrix decomposition
Optimization
Parallel algorithms
Partitioning
Partitioning algorithms
Performance analysis
Scalability
Software libraries
ISSN:1045-9219
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:This article presents various data redistribution methods for block-partitioned linear algebra algorithms operating on dense matrices that are distributed in a block-cyclic fashion. Because the algorithmic partitioning unit and the distribution blacking factor are most often chosen to be equal, severe alignment restrictions are induced on the operands, and optimal values with respect to performance are architecture dependent. The techniques presented in this paper redistribute data "on the fly," so that the user's data distribution blocking factor becomes independent from the architecture dependent algorithmic partitioning. These techniques are applied to the matrix-matrix multiplication operation. A performance analysis along with experimental results shows that alignment restrictions can then be removed and that high performance can be maintained across platforms independently from the user's data distribution blocking factor.
Bibliografia:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:1045-9219
DOI:10.1109/71.819944