Analysing value substitution and confidence estimation for value prediction

Value Prediction is one of the newest techniques used to break down ILP limits. Despite being under continuous study during the last few years, a few aspects related to this emerging technique remain unanalysed in depth. Exhaustively investigated in the context of control speculation, confidence est...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of systems architecture Jg. 47; H. 5; S. 459 - 475
Hauptverfasser: Piñuel, Luis, Moreno, Rafael A, Tirado, Francisco
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Amsterdam Elsevier B.V 01.05.2001
Elsevier
Elsevier Sequoia S.A
Schlagworte:
Applied sciences
Compound structure devices
Computer science; control theory; systems
Confidence estimation
Control speculation
Electronics
Exact sciences and technology
Logic programming
Programming theory
Semiconductor electronics. Microelectronics. Optoelectronics. Solid state devices
Simulation
Studies
Theoretical computing
Value analysis
Value prediction
Value prediction
Control speculation
Confidence estimation
Performance evaluation
Processor
Confidence
Program execution
Estimation
Prediction
Value analysis
Speculation
ISSN:1383-7621, 1873-6165
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Value Prediction is one of the newest techniques used to break down ILP limits. Despite being under continuous study during the last few years, a few aspects related to this emerging technique remain unanalysed in depth. Exhaustively investigated in the context of control speculation, confidence estimation has usually played a secondary role on value prediction and speculation. Closely linked to confidence estimation, value substitution also represents a relegated subject of research. This paper is focussed on analysing, in an isolated way, the respective impact on predictor performance of both confidence estimation and value substitution mechanisms. By using detailed pipeline-level simulations, we prove that improvements in these mechanisms are as important as reducing the predictor aliasing or even improving the prediction model.
Bibliographie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ISSN:1383-7621
1873-6165
DOI:10.1016/S1383-7621(01)00006-6