DocumentCode :
2344421
Title :
A Study of Dynamic Branch Predictors: Counter versus Perceptron
Author :
Ho, C.Y. ; Chong, K.F. ; Yau, C.H. ; Fong, Anthony S S
Author_Institution :
Dept. of Electron. Eng., City Univ. of Hong Kong
fYear :
2007
fDate :
2-4 April 2007
Firstpage :
528
Lastpage :
536
Abstract :
As the instruction issue rate and depth of pipelining increase, branch prediction is considered as a performance hurdle for modern processors. Extremely high branch prediction accuracy is essential to deliver their potential performance. Many strategies have been investigated to improve the dynamic branch prediction. Among them, those using 2-bit saturating up-down counter and those using perceptron have dominated over other predictors. This paper presents comparison between counter-based branch predictor and perceptron-based branch predictor and discusses the future developments of dynamic branch prediction
Keywords :
computer architecture; perceptrons; computer architecture; counter-based branch predictor; dynamic branch predictors; neural network; perceptron-based branch predictor; perceptrons; up-down counter; Accuracy; Artificial neural networks; Computer architecture; Counting circuits; History; Information technology; Microarchitecture; Neural networks; Pipeline processing; Shift registers;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Information Technology, 2007. ITNG '07. Fourth International Conference on
Conference_Location :
Las Vegas, NV
Print_ISBN :
0-7695-2776-0
Type :
conf
DOI :
10.1109/ITNG.2007.22
Filename :
4151738
Link To Document :
بازگشت