Performance evaluation of SUPRENUM for the LINPACK benchmark (Q1186186)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Performance evaluation of SUPRENUM for the LINPACK benchmark |
scientific article; zbMATH DE number 36309
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Performance evaluation of SUPRENUM for the LINPACK benchmark |
scientific article; zbMATH DE number 36309 |
Statements
Performance evaluation of SUPRENUM for the LINPACK benchmark (English)
0 references
28 June 1992
0 references
The authors report on some performance measurements for the LINPACK DGFA ( LU factorization) and DGESL (solves system of linear equations) subroutines on the SUPREMUM 1-cluster system with 16 processors. The parallelization of LINPACK routines uses a block-column wrapped interleaving scheme for distribution of the matrix over the processors. The performance measurements were obtained for a \(1000\times 1000\) matrix with various numbers of processors.
0 references
parallel computer
0 references
LINPACK benchmark
0 references
performance measurements
0 references
LINPACK DGFA
0 references
LU factorization
0 references
DGESL
0 references
SUPREMUM 1-cluster system
0 references
block-column wrapped interleaving scheme
0 references
