Flexible High-Performance Matrix Multiply via a Self-Modifying Runtime Code (2001)
by
Greg Henry
| Citations: | 5 - 0 self |
BibTeX
@TECHREPORT{Henry01flexiblehigh-performance,
author = {Greg Henry},
title = {Flexible High-Performance Matrix Multiply via a Self-Modifying Runtime Code},
institution = {},
year = {2001}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
With the advent of architectures with multiple levels of memory hierarchies, the quest for high performance implementation of commonly used linear algebra operations has become progressively more complex. Fortunately, recent research has shown that a large class of such operations can be implemented in a relatively portable fashion provided a high-performance matrix-matrix multiplication routine is available for a given architecture.
Citations
| 44 | Exploiting functional parallelism on power2 to design high-performance numerical algorithms - Agarwal, Gustavson, et al. - 1994 |
