ESL Power Estimation Methodology

  • Stefan Schuermans
  • Rainer Leupers


Although power consumption is an important metric and the biggest potential for savings is at electronic system level, the standard tools do not provide power estimation capabilities. A flexible and largely automated method for creation of high-level power models is presented. It is based on the collection of information from the timed functional SystemC simulations. Furthermore, the only input information about power consumption is a reference power curve, which matches a workload that can also be executed on the electronic system level simulator. Taking the information recorded from a reference workload executed on a SystemC simulator and a matching reference power curve, the power estimation method is able to create a linear power model. Four approaches for computing this linear power model from the reference data are presented.


  1. 1.
    Ben-Israel A, Greville TNE (2003) Generalized inverses: theory and applications, 2nd edn. Springer, New YorkzbMATHGoogle Scholar
  2. 2.
    Benini L, Bogliolo A, Favalli M, Micheli GD (1998) Regression models for behavioral power estimation. IOS J Integr Comput-Aided Eng 5(2):95–106CrossRefGoogle Scholar
  3. 3.
    Benini L, Hodgson R, Siegel P (1998) System-level power estimation and optimization. In: International symposium on low power electronics and design, ISLPED’98. ACM, New York, pp 173–178. Google Scholar
  4. 4.
    Bro R, de Jong S (1997) A fast non-negativity-constrained least squares algorithm. J Chemometrics 11:393–401.<393::AID-CEM483>3.0.CO;2-L CrossRefGoogle Scholar
  5. 5.
    Cai L, Gajski D (2003) Transaction level modeling: an overview. In: First IEEE/ACM/IFIP international conference on hardware/software codesign and systems synthesis (IEEE Cat. No.03TH8721), pp 19–24.
  6. 6.
    Grötker T, Liao S, Martin G, Swan S (2002) System design with SystemC. Kluwer Academic Publishers, Dordrecht.
  7. 7.
    Grüttner K, Hartmann PA, Fandrey T, Hylla K, Lorenz D, Stattelmann S, Sander B, Bringmann O, Nebel W, Rosenstiel W (2014) An ESL timing and power estimation and simulation framework for heterogeneous SoCs. In: International conference on embedded computer systems: architectures, modeling, and simulation, pp 181–190.
  8. 8.
    IEEE SystemC (2012) IEEE standard for standard SystemC language reference manual.
  9. 9.
    Intel Docea (2017) Intel Docea power and thermal modeling and simulation solutions. Retrieved 26 March 2017
  10. 10.
    Lawson CL, Hanson RJ (1995) Solving least squares problems. Soc Ind Appl Math
  11. 11.
    Martin G, Bailey B, Piziali A (2007) ESL design and verification. Morgan Kaufmann, BurlingtonGoogle Scholar
  12. 12.
    Schürmans S (2018) Power estimation on electronic system level using linear power models. Dissertation, RWTH Aachen University, Aachen.
  13. 13.
    Streubühr M, Rosales R, Hasholzner R, Haubelt C, Teich J (2011) ESL power and performance estimation for heterogeneous MPSoCs using SystemC. In: Forum on specification and design languages, pp 1–8Google Scholar
  14. 14.
    Synopsys platform architect MCO (2016) Synopsys platform architect MCO. Retrieved 22 December 2016
  15. 15.
    Trabelsi C, Ben Atitallah R, Meftali S, Dekeyser JL, Jemai A (2011) A model-driven approach for hybrid power estimation in embedded systems design. EURASIP J Embed Syst; CrossRefGoogle Scholar
  16. 16.
    Veller Y, Matalon S (2010) Why you should optimize power at the ESL. Retrieved 13 August 2016

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Stefan Schuermans
    • 1
  • Rainer Leupers
    • 2
  1. 1.Silexica GmbHKölnGermany
  2. 2.Institute for Communication Technologies and Embedded Systems (ICE)RWTH Aachen UniversityAachenGermany

Personalised recommendations