ABSTRACT
Near threshold computing (NTC) through aggressive supply voltage scaling has the potential to significantly improve energy-efficiency. However, the increase in variation-induced timing errors is a major challenge in NTC. This can be addressed in the scope of approximate computing by selectively embracing non-important variation-induced timing errors. In this paper, we propose a framework to leverage the error tolerance potential of approximate computing for energy-efficient NTC designs. In our framework, statistical timing error analysis as well as structural and functional error propagation analysis is performed to identify the approximable portion of a design. Then, a mixed-timing logic synthesis is employed to improve energy-efficiency by embracing errors in the approximable portion of the design. Experimental results show that the proposed approach can improve the energy-efficiency of NTC designs by more than 30%.
- Ghasemzadeh et.al. Power-aware computing in wearable sensor networks: An optimal feature selection. Mobile Computing, 2015.Google Scholar
- Samie et.al. Iot technologies for embedded computing. dens, 2016.Google Scholar
- Gebregiorgis et.al. Maximizing energy efficiency in ntc by variation-aware microprocessor pipeline optimization. In ISLPED, 2016. Google ScholarDigital Library
- Gebregiorgis et.al. A cross-layer analysis of soft error, aging and process variation in near threshold computing. In DATE, 2016. Google ScholarDigital Library
- Han et.al. Approximate computing: An emerging paradigm for energy-efficient design. In ETS, 2013.Google Scholar
- Almurib et.al. Inexact designs for approximate low power addition by cell replacement. In DATE, 2016. Google ScholarDigital Library
- Chippa et.al. Scalable effort hardware design: exploiting algorithmic resilience for energy efficiency. In DAC, 2010. Google ScholarDigital Library
- Park et.al. Dynamic bit-width adaptation in dct: an approach to trade off image quality and computation energy. TVLSI, 2010. Google ScholarDigital Library
- Shim et.al. Reliable low-power digital signal processing via reduced precision redundancy. TVLSI, 2004. Google ScholarDigital Library
- Hegde et.al. Soft digital signal processing. TVLSI, 2001. Google ScholarDigital Library
- Gupta et.al. Low-power digital signal processing using approximate adders. TCAD, 2013. Google ScholarDigital Library
- Yang et.al. Approximate xor/xnor-based adders for inexact computing. In IEEE-NANO, 2013.Google Scholar
- Ajay et.al. Variable latency speculative addition: A new paradigm for arithmetic circuit design. In DATE, 2008.Google Scholar
- Kyaw et.al. Low-power high-speed multiplier for error-tolerant application. In EDSSC, 2010.Google Scholar
- Akturk et.al. Decoupled control and data processing for approximate near-threshold voltage computing. IEEE Micro, 2015.Google Scholar
- Dreslinski et.al. Near-threshold computing: Reclaiming moore's law through energy efficient integrated circuits. Proceedings IEEE, 2010.Google Scholar
- Karlin. A first course in stochastic processes. Tutorial, 2014.Google Scholar
- G. Asadi and M.B. Tahoori. An analytical approach for soft error rate estimation in digital circuits. In ISCAS, 2005.Google ScholarCross Ref
- H. Asadi and M.B. Tahoori. Soft error modeling and protection for sequential elements. In DFT, 2005. Google ScholarDigital Library
- Sampson et.al. Enerj: Approximate data types for safe and general low-power computation. In ACM SIGPLAN, 2011. Google ScholarDigital Library
Recommendations
Improving Performance under Process and Voltage Variations in Near-Threshold Computing Using 3D ICs
Near-threshold computing (NTC) circuits have been shown to offer significant energy efficiency and power benefits but with a huge performance penalty. This performance loss exacerbates if process and voltage variations are considered. In this article, ...
Yield-driven near-threshold SRAM design
ICCAD '07: Proceedings of the 2007 IEEE/ACM international conference on Computer-aided designVoltage scaling is desirable in SRAM to reduce energy consumption. However, commercial SRAM is prone to functional failures when Vdd is scaled. Several SRAM designs scale Vdd to 200--300mV to minimize energy per access, but these designs do not consider ...
Asymmetric Underlapped FinFETs for Near- and Super-Threshold Logic at Sub-10nm Technology Nodes
Special Issue on Nanoelectronic Circuit and System Design Methods for the Mobile Computing Era and Regular PapersExtending double-gate FinFET scaling to sub-10nm technology regime requires device-engineering techniques for countering the rise of direct source to drain tunneling (DSDT), edge direct tunneling (EDT) and short channel effects (SCE) that degrade FinFET ...
Comments