Reducing Branch Misprediction Penalty through Confidence Estimation

The goal of this Thesis is reducing the global penalty associated to branch mispredictions, in terms of both performance degradation and energy consumption, through the use of confidence estimation. The reduction of this global penalty has been achieved, firstly, by increasing the accuracy of branch...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Aragón, Juan Luis
Formato: Articulo Revision
Lenguaje:Inglés
Publicado: 2002
Materias:
Ciencias Informáticas
Branch Prediction Reversal Unit (BPRU)
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/9459
http://journal.info.unlp.edu.ar/wp-content/uploads/JCST-Apr03-TO1.pdf
Aporte de:
SEDICI (UNLP) de Universidad Nacional de La Plata
id I19-R120-10915-9459
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Branch Prediction Reversal Unit (BPRU)
spellingShingle Ciencias Informáticas
Branch Prediction Reversal Unit (BPRU)
Aragón, Juan Luis
Reducing Branch Misprediction Penalty through Confidence Estimation
topic_facet Ciencias Informáticas
Branch Prediction Reversal Unit (BPRU)
description The goal of this Thesis is reducing the global penalty associated to branch mispredictions, in terms of both performance degradation and energy consumption, through the use of confidence estimation. The reduction of this global penalty has been achieved, firstly, by increasing the accuracy of branch predictors, next, by reducing the time necessary to restore the processor from a mispredicted branch, and finally, by reducing the energy consumption due to the execution of incorrect instructions. All these proposals rely on the use of confidence estimation, a mechanism that assesses the quality of branch predictions by means of estimating the probability of a dynamic branch prediction to be correct or incorrect.
format Articulo
Revision
author Aragón, Juan Luis
author_facet Aragón, Juan Luis
author_sort Aragón, Juan Luis
title Reducing Branch Misprediction Penalty through Confidence Estimation
title_short Reducing Branch Misprediction Penalty through Confidence Estimation
title_full Reducing Branch Misprediction Penalty through Confidence Estimation
title_fullStr Reducing Branch Misprediction Penalty through Confidence Estimation
title_full_unstemmed Reducing Branch Misprediction Penalty through Confidence Estimation
title_sort reducing branch misprediction penalty through confidence estimation
publishDate 2002
url http://sedici.unlp.edu.ar/handle/10915/9459
http://journal.info.unlp.edu.ar/wp-content/uploads/JCST-Apr03-TO1.pdf
work_keys_str_mv AT aragonjuanluis reducingbranchmispredictionpenaltythroughconfidenceestimation
bdutipo_str Repositorios
_version_ 1764820491649941506

Ejemplares similares