Negative-Bias Temperature Instability seriously affects nanoscale circuits reliability and performance. Continuous stress and increasing operating temperatures lead to device degradation and long-term system unavailability. The opportunity to optimize the duty-cycle of the stress/recovery phases to reduce Vth degradation leads to innovative research of reliability-oriented resources allocation at architectural level. This work explores the impact of different allocation strategies on the processor degradation, through a novel estimation methodology. Experimental results show that the proposed NBTI-aware allocation strategy can guarantee from 10% and up to 30% lower degradation compared to classical strategies, under different operating scenarios and under process variability.
NBTI Mitigation in Microprocessor Designs
CORBETTA, SIMONE;FORNACIARI, WILLIAM
2012-01-01
Abstract
Negative-Bias Temperature Instability seriously affects nanoscale circuits reliability and performance. Continuous stress and increasing operating temperatures lead to device degradation and long-term system unavailability. The opportunity to optimize the duty-cycle of the stress/recovery phases to reduce Vth degradation leads to innovative research of reliability-oriented resources allocation at architectural level. This work explores the impact of different allocation strategies on the processor degradation, through a novel estimation methodology. Experimental results show that the proposed NBTI-aware allocation strategy can guarantee from 10% and up to 30% lower degradation compared to classical strategies, under different operating scenarios and under process variability.File | Dimensione | Formato | |
---|---|---|---|
GLSVLSI2012.pdf
Accesso riservato
Descrizione: camera ready
:
Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione
424.42 kB
Formato
Adobe PDF
|
424.42 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.