Towards Fine-Grained DVFS in Embedded Multi-core CPUs

  • Giuseppe Massari
  • Federico Terraneo
  • Michele Zanella
  • Davide Zoni
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10793)

Abstract

Dynamic Voltage and Frequency Scaling (DVFS) is the preferred actuator for power-performance policies, and its use is growing also for thermal management. DVFS implementations, especially for embedded platforms, have historically provided only few possible operating points, despite this may impair the optimality of the frequency selection for a given application. Moreover, with multiple policies making use of it, the rate at which frequency and voltage changes will occur in a given system is expected to increase. The work presented in this paper has a two-fold objective: first, to present a methodology to extend a DVFS driver with additional operating points and second, to measure the impact of DVFS transitions from the performance and energy consumption perspective. This contribution can thus help both operating system and run-time manager designers to implement more efficient policies, as well as device driver programmers and hardware designers to optimize the DVFS infrastructure. The proposed approach has been tested on a quad-core ARM Cortex-A9 CPU based development board.

Notes

Acknowledgments

This work was supported in part by the European Union funded project under the grant M2DC H2020-688201 (http://www.m2dc.eu/en/) and MANGO H2020-671668 (http://www.mango-project.eu/).

References

  1. 1.
    Taylor, M.: A landscape of the new dark silicon design regime. In: IEEE Micro, pp. 8–19, September 2013Google Scholar
  2. 2.
    Nikov, K., Nunez-Yanez, J.L., Horsnell, M.: Evaluation of hybrid run-time power models for the ARM Big.LITTLE architecture. In: IEEE International Conference on Embedded and Ubiquitous Computing (EUC), October 2015Google Scholar
  3. 3.
    Garcia, R.C., Chung, J.M., Jo, S.W., Ha, T., Kyong, T.: Response time performance estimation in smartphones applying dynamic voltage & frequency scaling and completely fair scheduler. In: IEEE International Symposium on Consumer Electronics (ISCE), pp. 1–2, June 2014Google Scholar
  4. 4.
    Kwak, J., Choi, O., Chong, S., Mohapatra, P.: Dynamic speed scaling for energy minimization in delay-tolerant smartphone applications. In: IEEE Conference on Computer Communications (INFOCOM), pp. 2292–2300, April 2014Google Scholar
  5. 5.
    Park, J.G., Hsieh, C.Y., Dutt, N., Lim, S.S.: Quality-aware mobile graphics workload characterization for energy-efficient DVFS design. In: IEEE Symposium on Embedded Systems for Real-time Multimedia (ESTIMedia), October 2014Google Scholar
  6. 6.
    Srinivasan, S., Kurella, N., Koren, I., Kundu, S.: Dynamic reconfiguration vs. DVFS: a comparative study on power efficiency of processors. In: International Conference on VLSI Design and International Conference on Embedded Systems (VLSID), pp. 563–564, January 2016Google Scholar
  7. 7.
    Egilmez, B., Memik, G., Ogrenci-Memik, S., Ergin, O.: User-specific skin temperature-aware DVFS for smartphones. In: Design, Automation Test in Europe Conference Exhibition (DATE), pp. 1217–1220, March 2015Google Scholar
  8. 8.
    Leva, A., Terraneo, F., Giacomello, I., Fornaciari, W.: Event-based power/performance-aware thermal management for high-density microprocessors. IEEE Trans. Control Syst. Technol. 26, 535–550 (2017)CrossRefGoogle Scholar
  9. 9.
    Eyerman, S., Eeckhout, L.: Fine-grained DVFS Using on-chip regulators. ACM Trans. Archit. Code Optim. 8, 1:1–1:24 (2011)CrossRefGoogle Scholar
  10. 10.
    Juan, D.C., Garg, S., Park, J., Marculescu, D.: Learning the optimal operating point for many-core systems with extended range voltage/frequency scaling. In: International Conference on Hardware/Software Codesign and System Synthesis (CODES + ISSS), pp. 1–10. IEEE (2013)Google Scholar
  11. 11.
    Pan, J., Yoshihara, T.: A fast lock phase-locked loop using a continuous-time phase frequency detector. In: IEEE Conference on Electron Devices and Solid-State Circuits, pp. 393–396, December 2007Google Scholar
  12. 12.
    Abadian, A., Lotfizad, M., Majd, N.E., Ghoushchi, M.B.G., Mirzaie, H.: A new low-power and low-complexity all digital PLL (ADPLL) in 180 nm and 32 nm. In: IEEE International Conference on Electronics, Circuits and Systems (2010)Google Scholar
  13. 13.
    Kim, W., Gupta, M.S., Wei, G.Y., Brooks, D.: System level analysis of fast, per-core DVFS using on-chip switching regulators. In: IEEE International Symposium on High Performance Computer Architecture, pp. 123–134, February 2008Google Scholar
  14. 14.
    Altieri, M., Lombardi, W., Puschini, D., Lesecq, S.: Coupled voltage and frequency control for DVFS management. In: International Workshop on Power and Timing Modeling, Optimization and Simulation (PATMOS), September 2013Google Scholar
  15. 15.
    Park, S., Park, J., Shin, D., Wang, Y., Xie, Q.: Accurate modeling of the delay and energy overhead of dynamic voltage and frequency scaling in modern microprocessors. IEEE Trans. Comput.-Aided Des. Integr. Circ. Syst. 32, 695–708 (2013)CrossRefGoogle Scholar
  16. 16.
    Terraneo, F., Zoni, D., Fornaciari, W.: A cycle accurate simulation framework for asynchronous NoC design. In: International Symposium on System-on-Chip, SoC 2013 (2013)Google Scholar
  17. 17.
    Begum, R., Werner, D., Hempstead, M., Prasad, G., Challen, G.: Energy-performance trade-offs on energy-constrained devices with multi-component DVFS. In: IEEE International Symposium on Workload Characterization (IISWC) (2015)Google Scholar
  18. 18.
    Tan, L., Chen, Z., Zong, Z., Li, D., Ge, R.: A2E: Adaptively aggressive energy efficient DVFS scheduling for data intensive applications. In: IEEE International Performance Computing and Communications Conference (IPCCC) (2013)Google Scholar
  19. 19.
    Ge, R., Feng, X., Feng, W.C., Cameron, K.W.: CPU MISER: a performance-directed, run-time system for power-aware clusters. In: International Conference on Parallel Processing (ICPP), p. 18, September 2007Google Scholar
  20. 20.
  21. 21.
    Leva, A., Terraneo, F., Fornaciari, W.: Event-based control as an enabler for high power density processors. In: International Conference on Event-based Control, Communication, and Signal Processing (EBCCSP), June 2016Google Scholar
  22. 22.
    Rodopoulos, D., Catthoor, F., Soudris, D.: Tackling performance variability due to RAS mechanisms with PID-controlled DVFS. IEEE Comput. Architect. Lett. 14, 156–159 (2015)CrossRefGoogle Scholar
  23. 23.
    Liu, Y., Yang, H., Dick, R.P., Wang, H., Shang, L.: Thermal vs energy optimization for DVFS-enabled processors in embedded systems. In: International Symposium on Quality Electronic Design (ISQED), pp. 204–209 (2007)Google Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Giuseppe Massari
    • 1
  • Federico Terraneo
    • 1
  • Michele Zanella
    • 1
  • Davide Zoni
    • 1
  1. 1.DEIB, Politecnico di MilanoMilanItaly

Personalised recommendations