Adaptive Strassen and ATLAS's DGEMM: a fast square-matrix multiply for modern high-performance systems

Strassen's algorithm has practical performance benefits for architectures with simple memory hierarchies, because it trades computationally expensive matrix multiplications (MM) with cheaper matrix additions (MA). However, it presents no advantages for high-performance architectures with deep m...

Full description

Saved in:
Bibliographic Details
Published in:Eighth International Conference on High Performance Computing in Asia Pacific Region : proceedings : 30 November - 3 December, Beijing, China pp. 8 pp. - 52
Main Authors: D'Alberto, P., Nicolau, A.
Format: Conference Proceeding
Language:English
Published: IEEE 2005
Subjects:
Adaptive coding
Computer architecture
Computer science
Equations
High performance computing
IEEE members
Kernel
Memory architecture
Personal digital assistants
Software packages
ISBN:9780769524863, 0769524869
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Strassen's algorithm has practical performance benefits for architectures with simple memory hierarchies, because it trades computationally expensive matrix multiplications (MM) with cheaper matrix additions (MA). However, it presents no advantages for high-performance architectures with deep memory hierarchies, because MAs exploit limited data reuse. We present an easy-to-use adaptive algorithm combining Strassen's recursion and high-tuned version of ATLAS MM. In fact, we introduce a last step in the ATLAS-installation process that determines whether Strassen's may achieve any speedup. We present a recursive algorithm achieving up to 30% speed-up versus ATLAS alone. We show experimental results for 14 different systems
ISBN:9780769524863
0769524869
DOI:10.1109/HPCASIA.2005.18