THE COUNTERFLOW PIPELINE PROCESSOR ARCHITECTURE

被引：60
作者：
SPROULL, RF [1 ]
SUTHERLAND, IE [1 ]
MOLNAR, CE [1 ]
机构：
[1] WASHINGTON STATE UNIV,INST BIOMED COMP,PULLMAN,WA 99164
来源：
IEEE DESIGN & TEST OF COMPUTERS | 1994年 / 11卷 / 03期
关键词：
D O I：
10.1109/MDT.1994.303847
中图分类号：
TP3 [计算技术、计算机技术];
学科分类号：
0812 ;
摘要：
In a counterflow pipeline processor, instructions and results flow in opposite directions through a pipeline, interacting as they pass. Although detailed simulations of a complete processor design are not yet available, the architecture promises regularity in chip layout, local control to avoid performance limitations, and simplicity that may lead to provably correct designs. Moreover, CFPP designs allow asynchronous implementations more readily than conventional pipeline designs.
引用
收藏
页码：48 / 59
页数：12
相关论文
共 11 条
[1]  
BRUNVAND EL, 1989, P ICCAD
[2]  
BRUNVAND EL, 1987, CMUCS87119 CARN U CO
[3]  
CLARK WA, 1974, COMPUTERS BIOMEDICAL, V4, P45
[4]  
DILL DL, 1989, TRACE THEORY AUTOMAT
[5]  
HENNESSY JL, 1990, COMPUTER ARCHITECTUR, P261
[6]  
Hoare C.A.R., 1985, COMMUNICATING SEQUEN
[7]  
Mead C., 1980, INTRO VLSI SYSTEMS
[8]   THE METAFLOW ARCHITECTURE [J].
POPESCU, V ;
SCHULTZ, M ;
SPRACKLEN, J ;
GIBSON, G ;
LIGHTNER, B ;
ISAMAN, D .
IEEE MICRO, 1991, 11 (03) :10-&
[9]  
POPESCU V, 1991, IEEE MICRO, V11, P63
[10]  
SUTHERLAND I, 1991, ADV RES VLSI
12