Inexact computing using probabilistic circuits

Numerous computing applications can tolerate low error rates. In such applications, inexact approaches provide the ability to achieve significantly lower power. This work demonstrates the power-error trade-offs that can be achieved. Using probabilistic modeling in sub-50-nm silicon transistor technology, the relationship between statistical uncertainties and errors are elucidated for different configurations and topologies and the trade-offs quantified. Gate-level implementation of the probabilistic CMOS logic is validated by circuit simulations of a commercial 45-nm SOI CMOS process technology. Using a practical ALU architecture where voltages can be scaled from most significant to least significant bit blocks as an example, the potential benefits of this technique are shown. A calculation error of 10−6, an error rate quite tolerable for many computational tasks, is shown to be possible with a total power reduction of more than 40%.

[1]  R. A. WEALE.  Limits of Human Vision , 1961, Nature.

[2]  Nam Sung Kim,et al.  Energy-Efficient and Metastability-Immune Timing-Error Detection and Instruction-Replay-Based Recovery Circuits for Dynamic-Variation Tolerance , 2008, 2008 IEEE International Solid-State Circuits Conference - Digest of Technical Papers.

[3]  Sandip Tiwari,et al.  Scale changes in electronics: Implications for nanostructure devices for logic and memory and beyond , 2013 .

[4]  David Blaauw,et al.  A Power-Efficient 32 bit ARM Processor Using Timing-Error Detection and Correction for Transient-Error Tolerance and Adaptation to PVT Variation , 2011, IEEE Journal of Solid-State Circuits.

[5]  Anil Kumar,et al.  Statistical advantages of intrinsic channel fully depleted SOI MOSFETs over bulk MOSFETs , 2011, 2011 IEEE Custom Integrated Circuits Conference (CICC).

[6]  Sanjay Pant,et al.  A self-tuning DVS processor using delay-error detection and correction , 2005, IEEE Journal of Solid-State Circuits.

[7]  Ram Krishnamurthy,et al.  High-Performance Energy-Efficient Dual-Supply ALU Design , 2006 .

[8]  Jaeyoon Kim,et al.  Adaptive Circuit Design Using Independently Biased Back-Gated Double-Gate MOSFETS , 2012, IEEE Transactions on Circuits and Systems I: Regular Papers.

[9]  David Blaauw,et al.  Razor II: In Situ Error Detection and Correction for PVT and SER Tolerance , 2008, 2008 IEEE International Solid-State Circuits Conference - Digest of Technical Papers.

[10]  Steven M. Nowick,et al.  ACM Journal on Emerging Technologies in Computing Systems , 2010, TODE.

[11]  Michael Nicolaidis.  Time redundancy based soft-error tolerance to rescue nanometer technologies , 1999, Proceedings 17th IEEE VLSI Test Symposium (Cat. No.PR00146).

[12]  Douglas L. Jones,et al.  Computation as Estimation: A General Framework for Robustness and Energy Efficiency in SoCs , 2010, IEEE Transactions on Signal Processing.

[13]  B. Jagannathan,et al.  Record RF performance of 45-nm SOI CMOS Technology , 2007, 2007 IEEE International Electron Devices Meeting.

[14]  K.-U. Stein.  Noise-induced error rate as limiting factory for energy per operation in digital ICs , 1977 .

[15]  David Blaauw,et al.  Razor: circuit-level correction of timing errors for low-power operation , 2004, IEEE Micro.

[16]  Mark Horowitz,et al.  Clustered voltage scaling technique for low-power design , 1995, ISLPED '95.

[17]  Krishna V. Palem,et al.  Ultra-low energy computing with noise: Energy performance probability , 2006, IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures (ISVLSI'06).

[18]  Jaeyoon Kim,et al.  Inexact computing for ultra low-power nanometer digital circuit design , 2011, 2011 IEEE/ACM International Symposium on Nanoscale Architectures.

[19]  Jan M. Rabaey,et al.  Low Power Design Essentials , 2009, Series on Integrated Circuits and Systems.

[20]  Josef F. Bille.  The Limits of Human Vision , 2014 .

[21]  David M. Bull,et al.  RazorII: In Situ Error Detection and Correction for PVT and SER Tolerance , 2009, IEEE Journal of Solid-State Circuits.

[22]  Takashi Ishikawa,et al.  A low-power design method using multiple supply voltages , 1997, Proceedings of 1997 International Symposium on Low Power Electronics and Design.

[23]  Uming Ko,et al.  SmartReflex Power and Performance Management Technologies for 90 nm, 65 nm, and 45 nm Mobile Application Processors , 2010, Proceedings of the IEEE.

[24]  H. Vincent Poor,et al.  An Introduction to Signal Detection and Estimation , 1994, Springer Texts in Electrical Engineering.

[25]  H. Vincent Poor,et al.  An introduction to signal detection and estimation (2nd ed.) , 1994 .

[26]  Edward J. Nowak,et al.  Maintaining the benefits of CMOS scaling when scaling bogs down , 2002, IBM J. Res. Dev..