Michael Black
Applying Perceptrons to Speculation in Computer Architecture
Neural Networks in Future Microprocessors
Buch
Revision with unchanged content. Modern microprocessors make use of speculation, or predictions about future program behavior, to optimize the execution of programs. Perceptrons are simple neural networks that can be highly useful in speculation for their ability to examine larger quantities of available data than more commonly used approaches, and identify which data lead to accurate results. This work first studies how perceptrons can be made to predict accurately when they directly replace the traditional pattern table predictor. Different training methods, perceptron topologies, and interference reduction strategies are evaluated. Perce…
Mehr
Beschreibung
Revision with unchanged content. Modern microprocessors make use of speculation, or predictions about future program behavior, to optimize the execution of programs. Perceptrons are simple neural networks that can be highly useful in speculation for their ability to examine larger quantities of available data than more commonly used approaches, and identify which data lead to accurate results. This work first studies how perceptrons can be made to predict accurately when they directly replace the traditional pattern table predictor. Different training methods, perceptron topologies, and interference reduction strategies are evaluated. Perceptrons are then applied to two speculative applications: data value prediction and dataflow critical path prediction. Several novel perceptron-based prediction strategies are proposed for each application that can take advantage of a wider scope of past data in making predictions than previous predictors could. These predictors are evaluated against local table-based approaches on a custom cycle-accurate processor simulator, and are shown on average to have both superior accuracy and higher instruction-per-cycle performance. This work is addressed to computer architects and computer engineering researchers.
CHF 92.00
Preise inkl. MwSt. und Versandkosten (Portofrei ab CHF 40.00)
V105:
Folgt in ca. 15 Arbeitstagen
Produktdetails
- ISBN: 978-3-639-41699-2
- EAN: 9783639416992
- Produktnummer: 36361558
- Verlag: AV Akademikerverlag
- Sprache: Englisch
- Erscheinungsjahr: 2012
- Seitenangabe: 256 S.
- Masse: H22.0 cm x B15.0 cm x D1.5 cm 399 g
- Abbildungen: Paperback
- Gewicht: 399
Über den Autor
earned his Ph.D. in Electrical Engineering at the University of Maryland, College Park. He is currently an Assistant Professor of Computer Science at American University.
98 weitere Werke von Michael Black:
Neural Networks in Future Microprocessors
Ebook (EPUB Format)
CHF 6.45
Neural Networks in Future Microprocessors
Ebook (EPUB Format)
CHF 11.60
Neural Networks in Future Microprocessors
Ebook (EPUB Format)
CHF 5.00
Neural Networks in Future Microprocessors
Ebook (PDF Format)
CHF 58.05
Neural Networks in Future Microprocessors
Ebook (PDF Format)
CHF 94.50
Neural Networks in Future Microprocessors
Ebook (PDF Format)
CHF 308.50
Bewertungen
0 von 0 Bewertungen
Anmelden
Keine Bewertungen gefunden. Seien Sie der Erste und teilen Sie Ihre Erkenntnisse mit anderen.