ABSTRACT
Several recently proposed design techniques leverage the inherent error resilience of applications for improved efficiency (energy or performance). Hardware and software systems that are thus designed may be viewed as "scalable effort systems", since they offer the capability to modulate the effort that they expend towards computation, thereby allowing for tradeoffs between output quality and efficiency.
We propose the concept of Dynamic Effort Scaling (DES), which refers to dynamic management of the control knobs that are exposed by scalable effort systems. We argue the need for DES by observing that the degree of resilience often varies significantly across applications, across datasets, and even within a dataset. We propose a general conceptual framework for DES by formulating it as a feedback control problem, wherein the scaling mechanisms are regulated with the goal of maintaining output quality within a certain specified limit. We present an implementation of Dynamic Effort Scaling in the context of a scalable-effort processor for Support Vector Machines, and evaluate it under various application scenarios and data sets. Our results clearly demonstrate the benefits of the proposed approach --- statically setting the scaling mechanisms leads to either significant error overshoot or significant opportunities for energy savings left on the table unexploited. In contrast, DES is able to effectively regulate the output quality while maximally exploiting the time-varying resiliency in the workload.
- V. Wong and M. Horowitz. Soft error resilience of probabilistic inference applications. In Proc. SELSE, 2006.Google Scholar
- N. R. Shanbhag, R. A. Abdallah, R. Kumar, and D. L. Jones. Stochastic computation. In Proc. DAC, pages 859--864, 2010. Google ScholarDigital Library
- S. T. Chakradhar and A. Raghunathan. Best-effort Computing: Re-thinking Parallel Software and Hardware. In Proc. DAC, pages 865--870, 2010. Google ScholarDigital Library
- M. A. Breuer. Hardware that produces bounded rather than exact results. In Proc. DAC, pages 871--876, 2010. Google ScholarDigital Library
- A. P. Chandrakasan and R. W. Brodersen, editors. Low-Power CMOS Design. Wiley-IEEE Press, 1st edition, 1997. Google ScholarDigital Library
- L. Benini and G. DeMicheli. Dynamic Power Management: Design Techniques and CAD Tools. Kluwer Academic Publishers, Norwell, MA, USA, 1998. Google ScholarDigital Library
- R. Hegde and N. R. Shanbhag. Energy-efficient signal processing via algorithmic noise-tolerance. In Proc. ISLPED, pages 30--35, 1999. Google ScholarDigital Library
- Naresh Shanbhag. Reliable and energy-efficient digital signal processing. In Proc. DAC, pages 830--835, 2002. Google ScholarDigital Library
- R. Hegde and N. R. Shanbhag. A low-power digital filter IC via soft DSP. In Proc. CICC, pages 309--312, 2001.Google ScholarCross Ref
- G. V. Varatkar and N. R. Shanbhag. Error-resilient motion estimation architecture. IEEE Trans. VLSI Systems, 16(10):1399--1412, 2008. Google ScholarDigital Library
- Tong-Yu Hsieh, Kuen-Jong Lee, and M. A. Breuer. An error rate based test methodology to support error-tolerance. Reliability, IEEE Transactions on, 57(1):204--214, 2008.Google Scholar
- Z. Jiang and S. K. Gupta. An ATPG for threshold testing: Obtaining acceptable yield in future processes. In Proc. ITC, pages 824--833, 2002. Google ScholarDigital Library
- K. V. Palem. Energy aware algorithm design via probabilistic computing: From algorithms and models to Moore's law and novel (semiconductor) devices. In Proc. CASES, pages 113--116, 2003. Google ScholarDigital Library
- K. V. Palem, N. B. Chakrapani, Z. M. Kedem, A. Lingamneni, and K. K. Muntimadugu. Sustaining Moore's law in embedded computing through probabilistic and approximate design: Retrospects and prospects. In Proc. CASES, pages 1--10, 2009. Google ScholarDigital Library
- D. Mohapatra, G. Karakonstantis, and K. Roy. Significance driven computation: A voltage-scalable, variation-aware, quality-tuning motion estimator. In Proc. ISLPED, pages 195--200, 2009. Google ScholarDigital Library
- V. K. Chippa, D. Mohapatra, A. Raghunathan, K. Roy, and S. T. Chakradhar. Scalable effort hardware design: Exploiting algorithmic resilience for energy efficiency. In Proc. DAC, pages 555--560, 2010. Google ScholarDigital Library
- L. Leem, H. Cho, J. Bau, Q. A. Jacobson, and S. Mitra. ERSA: Error resilient system architecture for probabilistic applications. In Proc. DATE, pages 1560--1565, 2010. Google ScholarDigital Library
- W. Baek and T. M. Chilimbi. Green: A framework for supporting energy-conscious programming using controlled approximation. Proc. PLDI, pages 198--209, 2010. Google ScholarDigital Library
- Y. V. Ivanov and C. J. Bleakley. Dynamic complexity scaling for real-time H.264/AVC video encoding. In Proc. MULTIMEDIA, pages 962--970, 2007. Google ScholarDigital Library
- M. Shafique, L. Bauer, and J. Henkel. enBudget: A run-time adaptive predictive energy-budgeting scheme for energy-aware motion estimation in H.264/MPEG-4 AVC video encoder. In Proc. DATE, pages 1725--1730, 2010. Google ScholarDigital Library
- V. N. Vapnik. The nature of statistical learning theory. Springer-Verlag New York, Inc., New York, NY, USA, 1995. Google ScholarDigital Library
- Mohapatra Debabrata, Chippa Vinay, Raghunathan Anand, and Roy and Kaushik. Design of voltage scalable metafunctions for multimedia, recoginition and mining applications. In DATE, 2011.Google Scholar
- D. Ernst et al. Razor: A low-power pipeline based on circuit-level timing speculation. In Proc. MICRO, pages 7--18, 2003. Google ScholarDigital Library
- K. J. Astrom and T. Hagglund, editors. PID Controllers: Theory, Design, and Tuning. ISA, 2nd edition, 1995.Google Scholar
- J. Park, D. Shin, N. Chang, and M. Pedram. Accurate modeling and calculation of delay and energy overheads of dynamic voltage scaling in modern high-performance microprocessors. In Proc. ISLPED, pages 419--424, 2010. Google ScholarDigital Library
- Milde. www.nec-labs.com.Google Scholar
Index Terms
Dynamic effort scaling: managing the quality-efficiency tradeoff
Recommendations
Scalable effort hardware design: exploiting algorithmic resilience for energy efficiency
DAC '10: Proceedings of the 47th Design Automation ConferenceAlgorithms from several interesting application domains exhibit the property of inherent resilience to "errors" from extrinsic or intrinsic sources, offering entirely new avenues for performance and power optimization by relaxing the conventional ...
Managing the Quality vs. Efficiency Trade-off Using Dynamic Effort Scaling
Special Section on Probabilistic Embedded ComputingSeveral current and emerging applications do not have a unique result for a given input; rather, functional correctness is defined in terms of output quality. Recently proposed design techniques exploit the inherent resilience of such applications and ...
Sleep-Mode Voltage Scaling: Enabling SRAM Data Retention at Ultra-Low Power in Embedded Microcontrollers
Special Issue on VIPES, Special Issue on ICESS2015 and Regular PapersIn heavily duty-cycled embedded systems, the energy consumed by the microcontroller in idle mode is often the bottleneck for battery lifetime. Existing solutions address this problem by placing the microcontroller in a low-power (sleep) mode when idle ...
Comments