Abstract
Future semiconductors no longer guarantee permanent deterministic operation. They are expected to show probabilistic behavior due to lowered voltages and shrinking structures.
Compared to radiation-induced errors, probabilistic systems face increased error frequencies leading to unexpected bit-flips. Approaches like probabilistic CMOS provide methods to control error distributions which reduce the error probability in more significant bits. However, instructions handling control flow or pointers still expect deterministic operation, thus requiring a classification to identify these instructions.
We apply our transient error classification to probabilistic circuits using differing voltage distributions. Static analysis ensures that probabilistic effects only affect unreliable operations which accept a certain level of impreciseness, and that errors in probabilistic components will never propagate to critical operations.
To evaluate, we analyze robustness and quality-of-service of an H.264 video decoder. Using classification results, we map unreliable arithmetic operations onto probabilistic components of a simulated ARM-based architecture, while the remaining operations use deterministic components.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Benini, L., Bertozzi, D., Bogliolo, A., Menichelli, F., Olivieri, M.: MPARM: Exploring the Multi-Processor SoC Design Space with SystemC. The Journal of VLSI Signal Processing 41, 169–182 (2005)
Bhanu, A., Lau, M.S.K., Ling, K.-V., Mooney, V.J., Singh, A.: A more precise model of noise based CMOS errors. In: Proc. Intl. Symp. on Electronic Design, Test and Applications, pp. 99–102 (January 2010)
Budiu, M., Sakr, M., Walker, K., Goldstein, S.C.: BitValue Inference: Detecting and Exploiting Narrow Bitwidth Computations. In: Bode, A., Ludwig, T., Karl, W.C., Wismüller, R. (eds.) Euro-Par 2000. LNCS, vol. 1900, pp. 969–979. Springer, Heidelberg (2000)
Chakrapani, L.N.B., Muntimadugu, K.K., Lingamneni, A., George, J., Palem, K.V.: Highly energy and performance efficient embedded computing through approximately correct arithmetic. In: Proc. of CASES, pp. 187–196. ACM (2008)
Chin, B., Markstrum, S., Millstein, T., Palsberg, J.: Inference of User-Defined Type Qualifiers and Qualifier Rules. In: Sestoft, P. (ed.) ESOP 2006. LNCS, vol. 3924, pp. 264–278. Springer, Heidelberg (2006)
Dhoot, C., Mooney, V.J., Chau, L.P., Chowdhury, S.R.: Low power motion estimation with probabilistic computing. In: Proc. ISVLSI, pp. 176–181. IEEE (2011)
Engel, M., Schmoll, F., Heinig, A., Marwedel, P.: Unreliable yet useful – reliability annotations for data in cyber-physical systems. In: Proc. of the Workshop on Software Language Engineering for Cyber-Physical Systems, Berlin (October 2011)
Foster, J.S., Fähndrich, M., Aiken, A.: A theory of type qualifiers. In: Proc. of PLDI, pp. 192–203. ACM, New York (1999)
George, J., Marr, B., Akgul, B., Palem, K.: Probabilistic arithmetic and energy efficient embedded signal processing. In: Proc. of CASES, pp. 158–168. ACM (2006)
Gupta, A., Mandavalli, S., Mooney, V.J., Ling, K.-V., Basu, A., Johan, H., Tandianus, B.: Low power probabilistic floating point multiplier design. In: IEEE Computer Society Annual Symposium on VLSI (ISVLSI), pp. 182–187 (2011)
Heinig, A., Engel, M., Schmoll, F., Marwedel, P.: Improving transient memory fault resilience of an H.264 decoder. In: Proc. of ESTIMedia. IEEE (October 2010)
ICD e.V.: ICD-C Compiler framework, http://www.icd.de/es/icd-c/
Kedem, Z.M., Mooney, V.J., Muntimadugu, K.K., Palem, K.V.: An approach to energy-error tradeoffs in approximate ripple carry adders. In: Proc. Intl. Symposium on Low Power Electronics and Design (ISLPED), pp. 211–216 (August 2011)
Kedem, Z.M., Mooney, V.J., Muntimadugu, K.K., Palem, K., Devarasetty, A., Parasuramuni, P.D.: Optimizing energy to minimize errors in dataflow graphs using approximate adders. In: Proc. of CASES. pp. 177–186. ACM (2010)
Lau, M.S.K., Ling, K.-V., Bhanu, A., Mooney, V.J.: Error rate prediction for probabilistic circuits with more general structures. In: Proc. of the WS on Synthesis And System Integration of Mixed Information Technologies, pp. 220–225 (April 2010)
Li, X., Yeung, D.: Application-level correctness and its impact on fault tolerance. In: Proc. Symp. on High Performance Comp. Architecture, pp. 181–192 (2007)
Özer, E., Nisbet, A.P., Gregg, D.: A stochastic bitwidth estimation technique for compact and low-power custom processors. ACM TECS 7, 34:1–34:30 (2008)
Palem, K.V., Chakrapani, L.N.B., Kedem, Z.M., Lingamneni, A., Muntimadugu, K.K.: Sustaining moore’s law in embedded computing through probabilistic and approximate design: retrospects and prospects. In: Proc. of CASES. pp. 1–10. ACM (2009)
Palem, K.V.: Energy aware algorithm design via probabilistic computing: From algorithms and models to Moore’s Law and novel (semiconductor) devices. In: Proc. of CASES, pp. 113–116 (September 2003)
Palem, K.V.: Energy aware computing through probabilistic switching: A study of limits. IEEE Trans. Computers 54(9), 1123–1137 (2005)
Polian, I., Becker, B., Nakasato, M., Ohtake, S., Fujiwara, H.: Low-Cost Hardening of Image Processing Applications Against Soft Errors. In: Proc. of the Intl. Symp. on Defect and Fault-Tolerance in VLSI Systems, pp. 274–279. IEEE (2006)
Pouwelse, J., Langendoen, K., Sips, H.: Dynamic voltage scaling on a low-power microprocessor. In: Mobile Computing and Networking, pp. 251–259 (2001)
Sampson, A., Dietl, W., Fortuna, E., Gnanapragasam, D., Ceze, L., Grossman, D.: EnerJ: approximate data types for safe and general low-power computation. In: Proc. of PLDI, pp. 164–174. ACM, New York (2011)
Singh, A., Basu, A., Ling, K., Mooney, V.J.: Modeling multi-output filtering effects in PCMOS. In: Symp. on VLSI Design, Automation and Test, pp. 1–4 (April 2011)
Stephenson, M., Babb, J., Amarasinghe, S.: Bitwidth analysis with application to silicon compilation. In: Proc. of PLDI, pp. 108–120. ACM, New York (2000)
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Heinig, A., Mooney, V.J., Schmoll, F., Marwedel, P., Palem, K., Engel, M. (2012). Classification-Based Improvement of Application Robustness and Quality of Service in Probabilistic Computer Systems. In: Herkersdorf, A., Römer, K., Brinkschulte, U. (eds) Architecture of Computing Systems – ARCS 2012. ARCS 2012. Lecture Notes in Computer Science, vol 7179. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-28293-5_1
Download citation
DOI: https://doi.org/10.1007/978-3-642-28293-5_1
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-28292-8
Online ISBN: 978-3-642-28293-5
eBook Packages: Computer ScienceComputer Science (R0)