A Study of Dynamic Branch Predictors: Counter versus Perceptron

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...

Full description

Saved in:
Bibliographic Details
Main Authors: Ho, C.Y., Chong, K.F., Yau, C.H., Fong, A.S.S.
Format: Conference Proceeding
Language:English
Subjects:
Accuracy
Artificial neural networks
Computer architecture
Counting circuits
History
Information technology
Microarchitecture
Neural networks
Pipeline processing
Shift registers
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary: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
DOI:10.1109/ITNG.2007.22