Advertisement

Synthesis Time Reconfigurable Floating Point Unit for Transprecision Computing

  • Giulia StaziEmail author
  • Federica Silvestri
  • Antonio Mastrandrea
  • Mauro Olivieri
  • Francesco Menichelli
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 573)

Abstract

This paper presents the design and the implementation of a fully combinatorial floating point unit (FPU). The FPU can be reconfigured at implementation time in order to use an arbitrary number of bits for the mantissa and exponent, and it can be synthesized in order to support all IEEE-754 compliant FP formats but also non-standard FP formats, exploring the trade-off between precision (mantissa field), dynamic range (exponent field) and physical resources. This work is inspired by the consideration that, in modern low power embedded systems, the execution of floating point operations represents a significant contribution to energy consumption (up to 50% of the energy consumed by the CPU). In this scenario, the adoption of multiple FP formats, with a tunable number of bits for the mantissa and the exponent fields, is very interesting for reducing energy consumption and, simplifying the circuit, area and propagation delay. Adopting multiple FP formats on the same platform complies with the concept of transprecision computing, since it allows fine-grained control of approximation while meeting the required constraints on the precision of output results. The designed FPU has been tested in order to evaluate the correctness of all supported operations, and implemented on a Kintex-7 FPGA. Experimental results are provided, illustrating the impact and the benefits derived by the use of non-standard precision formats at circuit level.

Keywords

Floating point unit Low power consumption Approximate computing Transprecision computing 

References

  1. 1.
    Tagliavini, G., Mach, S., Rossi, D., Marongiu, A., Benini, L.: A transprecision floating-point platform for ultra-low power computing. In: Design, Automation & Test in Europe Conference & Exhibition (DATE), vol. 2018, pp. 1051–1056. IEEE (2018)Google Scholar
  2. 2.
    Han, J., Orshansky, M.: Approximate computing: An emerging paradigm for energy-efficient design. In: 18th IEEE European Test Symposium (ETS), vol. 2013, pp. 1–6. IEEE (2013)Google Scholar
  3. 3.
    Stazi, G., Menichelli, F., Mastrandrea, A., Olivieri, M.: Introducing approximate memory support in linux kernel. In: 2017 13th Conference on Ph.D. Research in Microelectronics and Electronics (PRIME), pp. 97–100. IEEE (2017)Google Scholar
  4. 4.
    Menichelli, F., Stazi, G., Mastrandrea, A., Olivieri, M.: An emulator for approximate memory platforms based on qemu. In: International Conference on Applications in Electronics Pervading Industry, Environment and Society, pp. 153–159. Springer (2016)Google Scholar
  5. 5.
    Stazi, G., Adani, L., Mastrandrea, A., Olivieri, M., Menichelli, F.: Impact of approximate memory data allocation on a h.264 software video encoder. In: International workshop on Approximate and Transprecision Computing on Emerging Technologies (ATCET). Springer (2018)Google Scholar
  6. 6.
    Malossi, A.C.I., Schaffner, M., Molnos, A., Gammaitoni, L., Tagliavini, G., Emerson, A., Tomás, A., Nikolopoulos, D.S., Flamand, E., Wehn, N.: The transprecision computing paradigm: concept, design, and applications. In: Design, Automation & Test in Europe Conference & Exhibition (DATE), vol. 2018, pp. 1105–1110. IEEE (2018)Google Scholar
  7. 7.
    Bailey, D.H., Yozo, H., Li, X.S., Thompson, B.: Arprec: an arbitrary precision computation package (2002)Google Scholar
  8. 8.
    Fousse, L., Hanrot, G., Lefèvre, V., Pélissier, P., Zimmermann, P.: Mpfr: a multiple-precision binary floating-point library with correct rounding. ACM Trans. Math. Softw. (TOMS) 33(2), 13 (2007)MathSciNetCrossRefGoogle Scholar
  9. 9.
    Kahan, W.: Ieee standard 754 for binary floating-point arithmetic. Lecture Notes on the Status of IEEE, vol. 754, no. 94720–1776, p. 11 (1996)Google Scholar
  10. 10.
    Graphics, M.: Modelsim-advanced simulation and debugging (2012)Google Scholar
  11. 11.
    Feist, T.: Vivado design suite. White Paper, vol. 5 (2012)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Giulia Stazi
    • 1
    Email author
  • Federica Silvestri
    • 1
  • Antonio Mastrandrea
    • 1
  • Mauro Olivieri
    • 1
  • Francesco Menichelli
    • 1
  1. 1.Department of Information Engineering, Electronics and Telecommunications (DIET)Sapienza University of RomeRomaItaly

Personalised recommendations