Algorithmic methodologies for ultra-efficient inexact architectures for sustaining technology scaling by Avinash Lingamneni, Kirthi Krishna Muntimadugu, Richard M. Karp, Krishna V. Palem, and Christian Piguet.
The following non-technical blurb caught my eye:
Researchers have unveiled an “inexact” computer chip that challenges the industry’s dogmatic 50-year pursuit of accuracy. The design improves power and resource efficiency by allowing for occasional errors. Prototypes unveiled this week at the ACM International Conference on Computing Frontiers in Cagliari, Italy, are at least 15 times more efficient than today’s technology.
[ads deleted]
The research, which earned best-paper honors at the conference, was conducted by experts from Rice University in Houston, Singapore’s Nanyang Technological University (NTU), Switzerland’s Center for Electronics and Microtechnology (CSEM) and the University of California, Berkeley.
“It is exciting to see this technology in a working chip that we can measure and validate for the first time,” said project leader Krishna Palem, who also serves as director of the Rice-NTU Institute for Sustainable and Applied Infodynamics (ISAID). “Our work since 2003 showed that significant gains were possible, and I am delighted that these working chips have met and even exceeded our expectations.” [From: Computing experts unveil superefficient ‘inexact’ chip which I saw in a list of links by Greg Linden.
Think about it. We are inexact and so are our semantics.
But we attempt to model our inexact semantics with increasingly exact computing platforms.
Does that sound like a modeling mis-match to you?
BTW, if you are interested in the details, see: Algorithmic methodologies for ultra-efficient inexact architectures for sustaining technology scaling
Abstract:
Owing to a growing desire to reduce energy consumption and widely anticipated hurdles to the continued technology scaling promised by Moore’s law, techniques and technologies such as inexact circuits and probabilistic CMOS (PCMOS) have gained prominence. These radical approaches trade accuracy at the hardware level for significant gains in energy consumption, area, and speed. While holding great promise, their ability to influence the broader milieu of computing is limited due to two shortcomings. First, they were mostly based on ad-hoc hand designs and did not consider algorithmically well-characterized automated design methodologies. Also, existing design approaches were limited to particular layers of abstraction such as physical, architectural and algorithmic or more broadly software. However, it is well-known that significant gains can be achieved by optimizing across the layers. To respond to this need, in this paper, we present an algorithmically well-founded cross-layer co-design framework (CCF) for automatically designing inexact hardware in the form of datapath elements. Specifically adders and multipliers, and show that significant associated gains can be achieved in terms of energy, area, and delay or speed. Our algorithms can achieve these gains with adding any additional hardware overhead. The proposed CCF framework embodies a symbiotic relationship between architecture and logic-layer design through the technique of probabilistic pruning combined with the novel confined voltage scaling technique introduced in this paper, applied at the physical layer. A second drawback of the state of the art with inexact design is the lack of physical evidence established through measuring fabricated ICs that the gains and other benefits that can be achieved are valid. Again, in this paper, we have addressed this shortcoming by using CCF to fabricate a prototype chip implementing inexact data-path elements; a range of 64-bit integer adders whose outputs can be erroneous. Through physical measurements of our prototype chip wherein the inexact adders admit expected relative error magnitudes of 10% or less, we have found that cumulative gains over comparable and fully accurate chips, quantified through the area-delay-energy product, can be a multiplicative factor of 15 or more. As evidence of the utility of these results, we demonstrate that despite admitting error while achieving gains, images processed using the FFT algorithm implemented using our inexact adders are visually discernible.
Why the link to the ACM Digital library or to the “unoffiical version” were not reported in any of the press stories I cannot say.