arrow
返回

Three-dimensional computational wavefronts for matrix product

delete1996-01-01
delete1
PRE
AI
J
JoAnn M. Paul *
M
Marlin H. Mickle
DOI:10.1016/0045-7906(95)00032-1delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
A three-dimensional wavefront array for matrix product with minimal block pipelining period of 1 is introduced and compared to existing systolic array architectures for matrix product. An optimal processor-time product of n(3) with cycles defined computationally by two operations is obtained when successive problem instances are considered. The 3-D architecture; is extensible and scalable, is cycle invariant (all respects), is node invariant (all respects), has minimal node complexity of one multiply and one addition per cycle, has unidirectional and local data forwarding in three dimensions, has 100% utilization of processing elements, and has a cycle-invariant one-to-one correspondence between input/output ports and input/output matrix elements.
Keyword:
computer architecture
pipelining
systolic array
throughput
matrix product
processor time
three-dimensional
wavefronts
parallel processing
latency

期刊

C
Computers and Electrical Engineering
IF:
4.9
论文数:
6.7K
被引数:
1.3W

机构

暂无机构信息
引用论文

引用论文

err分享
err收藏
err分享
err收藏
err分享
err收藏
Bis(N‐methylimidazole)‐Substituted Neutral Phthalocyanines {MIII(MeIm)2(Pc)·3–}0 (M = Al, Ga) Containing Radical Trianionic Phthalocyanine Macrocycles
err2016-08-06
err0
PREAI
errDmitri V. Konarev; Alexey V. Kuzmin; Salavat S. Khasanov; Akihiro Otsuka; Hideki Yamochi; Gunzi Saito; Rimma N. Lyubovskaya
err分享
err收藏
学者 查看更多内容