Optimizing the SVD Bidiagonalization Process for a Batch of Small Matrices

A challenging class of problems arising in many GPU applications, called batched problems, involves linear algebra operations on many small-sized matrices. We designed batched BLAS (Basic Linear Algebra Subroutines) routines, and in particular the Level-2 BLAS GEMV and the Level-3 BLAS GEMM routines...

Full description

Saved in:
Bibliographic Details
Published in:Procedia computer science Vol. 108; pp. 1008 - 1018
Main Authors: Dong, Tingxing, Haidar, Azzam, Tomov, Stanimire, Dongarra, Jack
Format: Journal Article
Language:English
Published: Elsevier B.V 2017
Subjects:
batched
Hardware accelerators
Singular Value Problems
two-sided factorization algorithms
batched
two-sided factorization algorithms
Hardware accelerators
Singular Value Problems
ISSN:1877-0509, 1877-0509
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A challenging class of problems arising in many GPU applications, called batched problems, involves linear algebra operations on many small-sized matrices. We designed batched BLAS (Basic Linear Algebra Subroutines) routines, and in particular the Level-2 BLAS GEMV and the Level-3 BLAS GEMM routines, to solve them. We proposed device functions and big-tile settings in our batched BLAS design. We adopted auto-tuning to optimize different instances of GEMV routines. We illustrated our batched BLAS approach to optimize batched bi-diagonalization progressively on a K40c GPU. The optimization techniques in this paper are applicable to the other two-sided factorizations as well.
ISSN:1877-0509
1877-0509
DOI:10.1016/j.procs.2017.05.237