Advertisement

A Lean, Low Power, Low Latency DRAM Memory Controller for Transprecision Computing

  • Chirag Sudarshan
  • Jan LappasEmail author
  • Christian Weis
  • Deepak M. Mathew
  • Matthias Jung
  • Norbert Wehn
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11733)

Abstract

Energy consumption is one of the major challenges for the advanced System on Chips (SoC). This is addressed by adopting heterogeneous and approximate computing techniques. One of the recent evolution in this context is transprecision computing paradigm. The idea of the transprecision computing is to consume adequate amount of energy for each operation by performing dynamic precision reduction. The impact of the memory subsystem plays a crucial role in such systems. Hence, the energy efficiency of a transprecision system can be further optimized by tailoring the memory subsystem to the transprecision computing. In this work, we present a lean, low power, low latency memory controller that is appropriate for transprecision methodology. The memory controller consumes an average power of 129.33 mW at a frequency of 500 MHz and has a total area of 4.71 mm2 for UMC 65 nm process.

Keywords

DRAM DDR3 Memory controller Transprecision PHY 

Notes

Acknowledgment

The project OPRECOMP acknowledges the financial support of the Future and Emerging Technologies (FET) programme within the European Unions Horizon 2020 research and innovation programme, under grant agreement No. 732631 (http://www.oprecomp.eu). This work was also supported by the Fraunhofer High Performance Center for Simulation- and Software-based Innovation.

References

  1. 1.
    Malossi, A.C.I., et al.: The transprecision computing paradigm: concept, design, and applications. In: 2018 Design, Automation Test in Europe Conference Exhibition (DATE), pp. 1105–1110, March 2018Google Scholar
  2. 2.
    Rossi, D., et al.: Energy-efficient near-threshold parallel computing: the PULPv2 cluster. IEEE Micro 37(5), 20–31 (2017)CrossRefGoogle Scholar
  3. 3.
    Weis, C., et al.: DRAMSpec: a high-level DRAM timing, power and area exploration tool. Int. J. Parallel Program. 45(6), 1566–1591 (2017)CrossRefGoogle Scholar
  4. 4.
    Jung, M., et al.: ConGen: an application specific dram memory controller generator. In: Proceedings of the Second International Symposium on Memory Systems, MEMSYS 2016, pp. 257–267. ACM, New York (2016)Google Scholar
  5. 5.
    Bhati, I., et al.: Flexible auto-refresh: enabling scalable and energy-efficient DRAM refresh reductions. In: Proceedings of the 42nd Annual International Symposium on Computer Architecture, pp. 235–246. ACM (2015)Google Scholar
  6. 6.
    Liu, J., et al.: RAIDR: retention-aware intelligent dram refresh. In: Proceedings of the 39th Annual International Symposium on Computer Architecture, ISCA 2012, pp. 1–12. IEEE Computer Society, Washington (2012)Google Scholar
  7. 7.
    Jung, M., et al.: Approximate computing with partially unreliable dynamic random access memory - approximate DRAM. In: Proceedings of the 53rd Annual Design Automation Conference, DAC 2016, pp. 100:1–100:4. ACM, New York (2016)Google Scholar
  8. 8.
    Lucas, J., et al.: Sparkk: quality-scalable approximate storage in DRAM. In: The Memory Forum, June 2014Google Scholar
  9. 9.
    Liu, S., et al.: Flikker: saving DRAM refresh-power through critical data partitioning. SIGPLAN Not. 46(3), 213–224 (2011)CrossRefGoogle Scholar
  10. 10.
    Jung, M.., et al.: Omitting refresh - a case study for commodity and wide I/O DRAMs. In: 1st International Symposium on Memory Systems, MEMSYS 2015, Washington, DC, USA, October 2015Google Scholar
  11. 11.
    Mathew, D.M., et al.: Using run-time reverse-engineering to optimize DRAM refresh. In: International Symposium on Memory Systems (MEMSYS17) (2017)Google Scholar
  12. 12.
    Jedec Solid State Technology Association. DDR3 SDRAM (JESD 79–3) (2012)Google Scholar
  13. 13.
    Cadence Inc.: Cadence® Denali® DDR Memory IP, October 2014. http://ip.cadence.com/ipportfolio/ip-portfolio-overview/memory-ip/ddr-lpddr. Accessed 18 Feb 2015
  14. 14.
    Synopsys, Inc.: DesignWare DDR IP (2015). http://www.synopsys.com/IP/ InterfaceIP/DDRn/Pages/. Accessed 18 Feb 2015
  15. 15.
    Fan, X., et al.: ESD protection circuit schemes for DDR3 DQ drivers. In: Electrical Overstress/Electrostatic Discharge Symposium Proceedings, pp. 1–6 (2010)Google Scholar
  16. 16.
    Yoo, C., et al.: A 1.8 V 700 Mb/s/pin 512 Mb DDR-II SDRAM with on-die termination and off-chip driver calibration. In: 2003 IEEE International Solid-State Circuits Conference, ISSCC 2003. Digest of Technical Papers, vol. 1, pp. 312–496, February 2003Google Scholar
  17. 17.
    Chen, S., et al.: An all-digital delay-locked loop for high-speed memory interface applications. In: Technical Papers of 2014 International Symposium on VLSI Design, Automation and Test, pp. 1–4, April 2014Google Scholar
  18. 18.
    De Caro, D.: Glitch-free NAND-based digitally controlled delay-lines. IEEE Trans. Very Large Scale Integr. (VLSI) Syst. 21(1), 55–66 (2013)CrossRefGoogle Scholar
  19. 19.
    Micron: DDR3 SDRAM System Power Calculator, July 2011. Accessed 03 July 2014Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Chirag Sudarshan
    • 1
  • Jan Lappas
    • 1
    Email author
  • Christian Weis
    • 1
  • Deepak M. Mathew
    • 1
  • Matthias Jung
    • 2
  • Norbert Wehn
    • 1
  1. 1.Technische Universität KaiserslauternKaiserslauternGermany
  2. 2.Fraunhofer Institute for Experimental Software Engineering (IESE)KaiserslauternGermany

Personalised recommendations