Fundamental Limits on Run-Time Power Management Algorithms for MPSoCs

  • Siddharth Garg
  • Diana Marculescu
  • Radu Marculescu


We consider the case of multiprocessor systems-on-chip (MPSoC) implemented using multiple voltage and frequency islands (VFIs) relying on fine-grained dynamic voltage and frequency scaling (DVFS) to reduce the system power dissipation. We present a framework to theoretically analyze the impact of three important technology driven constraints; (i) reliability-driven upper limits on the maximum supply voltage; (ii) inductive noise-driven constraints on the maximum rate of change of voltage/frequency; and (iii) the impact of manufacturing process variations on the performance of DVFS control for multiple VFI MPSoCs. The proposed analysis is general, in the sense that it is not bound to a specific DVFS control algorithm, but instead focuses on theoretically bounding the performance that any DVFS controller can possibly achieve. Our experimental results on real and synthetic benchmarks show that in the presence of reliability and temperature driven constraints on the maximum frequency and maximum frequency increment, any DVFS control algorithm will lose up to 87% performance in terms of the number of steps required to reach a reference steady state. In addition, increasing process variations can lead to up to 60% of fabricated chips being unable to meet the specified DVFS control specifications, irrespective of the DVFS algorithm used (the material in this paper is based on a prior work by the authors as described in Garg etal., ACM J Emerg Technol Comput Syst (JETC) 8(4):28, 2012; Technology-driven limits on DVFS controllability of multiple voltage-frequency island designs: a system-level perspective. In: Proceedings of the 46th IEEE/ACM design automation conference, San Francisco. IEEE, pp818–821, 2009).



Siddharth Garg acknowledges financial support from the Conseil de Recherches en Sciences Naturelles et en Genie du Canada (CRSNG) Discovery Grants program. Diana Marculescu and Radu Marculescu acknowledge partial support by the National Science Foundation (NSF) under grants CCF-0916752 and CNS-1128624.


  1. 1.
    Beigne E, Clermidy F, Lhermet H, Miermont S, Thonnart Y, Tran XT, Valentian A, Varreau D, Vivet P, Popon X et al (2009) An asynchronous power aware and adaptive NoC based circuit. IEEE J Solid-State Circuit 44(4):1167–1177CrossRefGoogle Scholar
  2. 2.
    Carta S, Alimonda A, Pisano A, Acquaviva A, Benini L (2007) A control theoretic approach to energy-efficient pipelined computation in MPSoCs. ACM Trans Embedded Comput Syst (TECS) 6(4):27–esGoogle Scholar
  3. 3.
    Clermidy F, Bernard C, Lemaire R, Martin J, Miro-Panades I, Thonnart Y, Vivet P, Wehn N (2010) MAGALI: a network-on-chip based multi-core system-on-chip for MIMO 4G SDR. In: Proceedings of the IEEE international conference on IC design and technology (ICICDT). Grenoble, France IEEE, pp 74–77Google Scholar
  4. 4.
    David R, Bogdan B, Marculescu R (2012) Dynamic power management for multi-cores: case study using the Intel SCC. In: Proceedings of VLSI SOC conference. Santa Cruz, CA IEEEGoogle Scholar
  5. 5.
    Dighe S, Vangal S, Aseron P, Kumar S, Jacob T, Bowman K, Howard J, Tschanz J, Erraguntla V, Borkar N et al (2010) Within-die variation-aware dynamic-voltage-frequency scaling core mapping and thread hopping for an 80-core processor. In: IEEE solid-state circuits conference digest of technical papers. San Francisco, CA IEEE, pp 174–175Google Scholar
  6. 6.
    Garg S, Marculescu D, Marculescu R, Ogras U (2009) Technology-driven limits on DVFS controllability of multiple voltage-frequency island designs: a system-level perspective. In: Proceedings of the 46th IEEE/ACM design automation conference, San Francisco, CA IEEE, pp 818–821Google Scholar
  7. 7.
    Garg S, Marculescu D, Marculescu R (2010) Custom feedback control: enabling truly scalable on-chip power management for MPSoCs. In: Proceedings of the 16th ACM/IEEE international symposium on low power electronics and design. Austin, TX ACM, pp 425–430Google Scholar
  8. 8.
    Garg S, Marculescu D, Marculescu R (2012) Technology-driven limits on runtime power management algorithms for multiprocessor systems-on-chip. ACM J Emerg Technol Comput Syst (JETC) 8(4):28Google Scholar
  9. 9.
    Goulding-Hotta N, Sampson J, Venkatesh G, Garcia S, Auricchio J, Huang PC, Arora M, Nath S, Bhatt V, Babb J et al (2011) The GreenDroid mobile application processor: an architecture for silicon’s dark future. IEEE Micro 31:86–95CrossRefGoogle Scholar
  10. 10.
    Jang W, Ding D, Pan DZ (2008) A voltage-frequency island aware energy optimization framework for networks-on-chip. In: Proceedings of the IEEE/ACM international conference on computer-aided design, San JoseGoogle Scholar
  11. 11.
    Keskin G, Li X, Pileggi L (2006) Active on-die suppression of power supply noise. In: Proceedings of the IEEE custom integrated circuits conference, San Jose. IEEE, pp 813–816Google Scholar
  12. 12.
    Kim W, Gupta MS, Wei GY, Brooks D (2008) System level analysis of fast, per-core DVFS using on-chip switching regulators. In: Proceedings of the 14th international symposium on high performance computer architecture. Salt Lake City, UT IEEE, pp 123–134Google Scholar
  13. 13.
    Kuo BC (1992) Digital control systems. Oxford University Press, New YorkGoogle Scholar
  14. 14.
    Marculescu D, Garg S (2006) System-level process-driven variability analysis for single and multiple voltage-frequency island systems. In: Proceedings of the 2006 IEEE/ACM international conference on computer-aided design. San Jose, CAGoogle Scholar
  15. 15.
    Mezhiba AV, Friedman EG (2004) Scaling trends of on-chip power distribution noise. IEEE Trans Very Large Scale Integr (VLSI) Syst 12(4):386–394CrossRefGoogle Scholar
  16. 16.
    Niyogi K, Marculescu D (2005) Speed and voltage selection for GALS systems based on voltage/frequency islands. In: Proceedings of the 2005 conference on Asia South Pacific design automation, ShanghaiGoogle Scholar
  17. 17.
    Ogras UY, Marculescu R, Marculescu D (2008) Variation-adaptive feedback control for networks-on-chip with multiple clock domains. In: Proceedings of the 45th annual conference on design automation. Annaheim, CAGoogle Scholar
  18. 18.
    Ravishankar C, Ananthanarayanan S, Garg S, Kennings A (2012) Analysis and evaluation of greedy thread swapping based dynamic power management for MPSoC platforms. In: Proceedings of the 13th international symposium on quality electronic design (ISQED), Santa Clara. IEEE, pp 617–624Google Scholar
  19. 19.
    Royden HL (1968) Real analysis. Macmillan, New YorkGoogle Scholar
  20. 20.
    Salihundam P, Jain S, Jacob T, Kumar S, Erraguntla V, Hoskote Y, Vangal S, Ruhl G, Borkar N (2011) A 2 Tb/s 6 ×4 mesh network for a single-chip cloud computer with DVFS in 45 nm CMOS. IEEE J Solid State Circuit 46(4):757–766CrossRefGoogle Scholar
  21. 21.
    Stanley-Marbell P, Marculescu D (2007) Sunflower: full-system, embedded microarchitecture evaluation. In: De Bosschere K, Kaeli D, Stenström P, Whalley D, Ungerer T (eds) High performance embedded architectures and compilers. Springer, Berlin, pp 168–182CrossRefGoogle Scholar
  22. 22.
    Truong D, Cheng W, Mohsenin T, Yu Z, Jacobson T, Landge G, Meeuwsen M, Watnik C, Mejia P, Tran A et al (2008) A 167-processor 65 nm computational platform with per-processor dynamic supply voltage and dynamic clock frequency scaling. In: Proceedings of the IEEE symposium on VLSI circuits. Honolulu, Hawaii IEEE, pp 22–23Google Scholar
  23. 23.
    Wang Y, Ma K, Wang X (2009) Temperature-constrained power control for chip multiprocessors with online model estimation. In: ACM SIGARCH computer architecture news, vol 37. ACM, pp 314–324Google Scholar
  24. 24.
    Wilkerson C, Gao H, Alameldeen AR, Chishti Z, Khellah M, Lu SL (2008) Trading off cache capacity for reliability to enable low voltage operation. In: ACM SIGARCH computer architecture news, vol 36. IEEE Computer Society, pp 203–214Google Scholar
  25. 25.
    Wu Q, Juang P, Martonosi M, Clark DW (2004) Formal online methods for voltage/frequency control in multiple clock domain microprocessors. ACM SIGOPS Oper Syst Rev 38(5): 248–259CrossRefGoogle Scholar
  26. 26.
    Zanini F, Atienza D, Benini L, De Micheli G (2009) Multicore thermal management with model predictive control. In: Proceedings of the European conference on circuit theory and design. Antalya, Turkey IEEE, pp 711–714Google Scholar
  27. 27.
    Zhao W, Cao Y (2006) New generation of predictive technology model for sub-45 nm design exploration. In: Proceedings of ISQED, San JoseGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2013

Authors and Affiliations

  • Siddharth Garg
    • 1
  • Diana Marculescu
    • 2
  • Radu Marculescu
    • 2
  1. 1.University of WaterlooWaterlooCanada
  2. 2.Carnegie Mellon UniversityPittsburghUSA

Personalised recommendations