A central component of modern computing is the idea that computation requiresdeterminism. Contrary to this belief, the primary contribution of this work shows thatuseful computation can be accomplished in an error-prone fashion. Focusing on low-powercomputing and the increasing push toward energy conservation, the work seeks to sacrificeaccuracy in exchange for energy savings.Probabilistic computing forms the basis for this error-prone computation by diverging from the requirement of determinism and allowing for randomness within computing.Implemented as probabilistic CMOS (PCMOS), the approach realizes enormous energy sav-ings in applications that require probability at an algorithmic level. Extending probabilisticcomputing to applications that are inherently deterministic, the biased voltage overscaling(BIVOS) technique presented here constrains the randomness introduced through PCMOS.Doing so, BIVOS is able to limit the magnitude of any resulting deviations and realizesenergy savings with minimal impact to application quality.Implemented for a ripple-carry adder, array multiplier, and finite-impulse-response (FIR)filter; a BIVOS solution substantially reduces energy consumption and does so with im-proved error rates compared to an energy equivalent reduced-precision solution. Whenapplied to H.264 video decoding, a BIVOS solution is able to achieve a 33.9% reduction inenergy consumption while maintaining a peak-signal-to-noise ratio of 35.0dB (compared to14.3dB for a comparable reduced-precision solution).While the work presented here focuses on a specific technology, the technique realizedthrough BIVOS has far broader implications. It is the departure from the conventionalmindset that useful computation requires determinism that represents the primary innovation of this work. With applicability to emerging and yet to be discovered technologies,BIVOS has the potential to contribute to computing in a variety of fashions.
【 预 览 】
附件列表
Files
Size
Format
View
Harnessing resilience: biased voltage overscaling for probabilistic signal processing