skip to main content
research-article

Dataflow Driven Partitioning of Machine Learning Applications for Optimal Energy Use in Batteryless Systems

Published:09 December 2022Publication History
Skip Abstract Section

Abstract

Sensing systems powered by energy harvesting have traditionally been designed to tolerate long periods without energy. As the Internet of Things (IoT) evolves toward a more transient and opportunistic execution paradigm, reducing energy storage costs will be key for its economic and ecologic viability. However, decreasing energy storage in harvesting systems introduces reliability issues. Transducers only produce intermittent energy at low voltage and current levels, making guaranteed task completion a challenge. Existing ad hoc methods overcome this by buffering enough energy either for single tasks, incurring large data-retention overheads, or for one full application cycle, requiring a large energy buffer. We present Julienning: an automated method for optimizing the total energy cost of batteryless applications. Using a custom specification model, developers can describe transient applications as a set of atomically executed kernels with explicit data dependencies. Our optimization flow can partition data- and energy-intensive applications into multiple execution cycles with bounded energy consumption. By leveraging interkernel data dependencies, these energy-bounded execution cycles minimize the number of system activations and nonvolatile data transfers, and thus the total energy overhead. We validate our methodology with two batteryless cameras running energy-intensive machine learning applications. Using a solar testbed, we replay real-world illuminance traces to experimentally demonstrate optimized batteryless execution with a transducer-to-application energy efficiency of 74.5%. Partitioning results demonstrate that compared to ad hoc solutions, our method can reduce the required energy storage by over 94% while only incurring a 0.12% energy overhead.

REFERENCES

  1. [1] Abdel-Hamid Ossama, Deng Li, and Yu Dong. 2013. Exploring convolutional neural network structures and optimization techniques for speech recognition. In Interspeech, Vol. 11. Citeseer, 7375.Google ScholarGoogle Scholar
  2. [2] Afanasov Mikhail, Bhatti Naveed Anwar, et al. 2020. Batteryless zero-maintenance embedded sensing at the Mithræum of circus maximus. In Proc. SenSys Conf.ACM, 368381.Google ScholarGoogle Scholar
  3. [3] Asad Hafiz Areeb, Wouters Erik Henricus, Bhatti Naveed Anwar, Mottola Luca, and Voigt Thiemo. 2020. On securing persistent state in intermittent computing. In Proc. ENSSys Workshop.Google ScholarGoogle Scholar
  4. [4] Balsamo Domenico, Weddell Alex S., Merrett Geoff V., Al-hashimi Bashir M., Brunelli Davide, and Benini Luca. 2015. Hibernus: Sustaining computation during intermittent supply for energy-harvesting systems. Embed. Syst. Lett. IEEE 7, 1 (2015). Google ScholarGoogle ScholarDigital LibraryDigital Library
  5. [5] Bhatti Naveed and Mottola Luca. 2016. Efficient state retention for transiently-powered embedded sensing. In Int. Conf. Embed. Wireless Sys. Netw.137148.Google ScholarGoogle Scholar
  6. [6] Bhatti Naveed Anwar, Alizai Muhammad Hamad, Syed Affan A., and Mottola Luca. 2016. Energy harvesting and wireless transfer in sensor network applications: Concepts and experiences. ACM Trans. Sensor Netw. (TOSN) 12, 3 (2016), 24.Google ScholarGoogle ScholarDigital LibraryDigital Library
  7. [7] Bhatti Naveed Anwar and Mottola Luca. 2017. HarvOS: Efficient code instrumentation for transiently-powered embedded sensing. In Proc. IPSN Conf.ACM, New York, NY. Google ScholarGoogle ScholarDigital LibraryDigital Library
  8. [8] Buchli Bernhard, Sutton Felix, Beutel Jan, and Thiele Lothar. 2014. Towards enabling uninterrupted long-term operation of solar energy harvesting embedded systems. In Proc. EWSN Conf. Springer, 6683.Google ScholarGoogle Scholar
  9. [9] Buettner Michael, Greenstein Ben, and Wetherall David. 2011. Dewdrop: An energy-aware runtime for computational RFID. In Proc. USENIX NSDI. 197210.Google ScholarGoogle Scholar
  10. [10] Colin Alexei and Lucia Brandon. 2016. Chain: Tasks and channels for reliable intermittent programs. In Proc. OOPSLA Conf.ACM. Google ScholarGoogle ScholarDigital LibraryDigital Library
  11. [11] Colin Alexei and Lucia Brandon. 2018. Termination checking and task decomposition for task-based intermittent programs. In Proc. Comp. Construct. Conf. ACM. Google ScholarGoogle ScholarDigital LibraryDigital Library
  12. [12] Colin Alexei, Ruppel Emily, and Lucia Brandon. 2018. A reconfigurable energy storage architecture for energy-harvesting devices. In Proc. ASPLOS Conf.ACM. Google ScholarGoogle ScholarDigital LibraryDigital Library
  13. [13] Daulby Tim, Savanth Anand, Weddell Alex S., and Merrett Geoff V.. 2020. Comparing NVM technologies through the lens of intermittent computation. In Proc. ENSSys Workshop. 7778.Google ScholarGoogle Scholar
  14. [14] Dijkstra Edsger W.. 1959. A note on two problems in connexion with graphs. Numer. Math. 1, 1 (Dec.1959), 269271. Google ScholarGoogle ScholarDigital LibraryDigital Library
  15. [15] FLIR Systems, Inc. 2018. Lepton Engineering Datasheet. FLIR Systems, Inc. Rev. 200.Google ScholarGoogle Scholar
  16. [16] Golonzka Oleg, Alzate J.-G., Arslan U., Bohr M., Bai P., Brockman J., Buford B., Connor C., Das N., Doyle B., et al. 2018. MRAM as embedded non-volatile memory solution for 22FFL FinFET technology. In 2018 IEEE International Electron Devices Meeting (IEDM). IEEE, 18–1.Google ScholarGoogle Scholar
  17. [17] Gomez Andres. 2020. On-demand communication with the batteryless mirocard: Demo abstract. In Proc. 18th Conf. Embed. Netw. Sens. Syst. 629630.Google ScholarGoogle Scholar
  18. [18] Gomez Andres et al. 2015. Reducing energy consumption in microcontroller-based platforms with low design margin co-processors. In Proc. DATE Conf.Google ScholarGoogle Scholar
  19. [19] Gomez Andres, Conti Francesco, and Benini Luca. 2018. Thermal image-based CNN’s for ultra-low power people recognition. In Proc. Comput. Frontiers Conf. ACM.Google ScholarGoogle Scholar
  20. [20] Gomez Andres, Sigrist Lukas, Magno Michele, Benini Luca, and Thiele Lothar. 2016. Dynamic energy burst scaling for transiently powered systems. In Proc. DATE Conf. EDA Consortium, 349354.Google ScholarGoogle Scholar
  21. [21] Gomez Andres, Sigrist Lukas, Schalch Thomas, Benini Luca, and Thiele Lothar. 2017. Wearable, energy-opportunistic vision sensing for walking speed estimation. In Proc. SAS Symp. IEEE, 16.Google ScholarGoogle Scholar
  22. [22] Gomez Andres, Sigrist Lukas, Schalch Thomas, Benini Luca, and Thiele Lothar. 2018. Efficient, long-term logging of rich data sensors using transient sensor nodes. ACM Trans. Embed. Comput. Syst. 17, 1 (2018). Google ScholarGoogle ScholarDigital LibraryDigital Library
  23. [23] Hager Pascal Alexander, Fatemi Hamed, Gyvez Jose Pineda de, and Benini Luca. 2017. A scan-chain based state retention methodology for IoT processors operating on intermittent energy. In Proc. DATE Conf. EDA Consortium.Google ScholarGoogle Scholar
  24. [24] Hester Josiah, Sitanayah Lanny, and Sorber Jacob. 2015. Tragedy of the Coulombs: Federating energy storage for tiny, intermittently-powered sensors. In Proc. SenSys Conf. ACM.Google ScholarGoogle Scholar
  25. [25] Hester Josiah, Storer Kevin, and Sorber Jacob. 2017. Timely execution on intermittently powered batteryless sensors. In Proc. SenSys Conf.Google ScholarGoogle Scholar
  26. [26] Hu Jingtong, Xue Chun Jason, Zhuge Qingfeng, Tseng Wei-Che, and Sha Edwin H.-M.. 2013. Write activity reduction on non-volatile main memories for embedded chip multiprocessors. ACM Trans. Embed. Comput. Syst. 12, 3 (April2013), Article 77, 27 pages. Google ScholarGoogle ScholarDigital LibraryDigital Library
  27. [27] Jackson Neal, Adkins Joshua, and Dutta Prabal. 2019. Capacity over capacitance for reliable energy harvesting sensors. In Proc. 18th Int. Conf. Inf. Process. Sensor Netw. 193204.Google ScholarGoogle Scholar
  28. [28] Jayakumar Hrishikesh, Raha Arnab, and Raghunathan Vijay. 2014. QUICKRECALL: A low overhead HW/SW approach for enabling computations across power cycles in transiently powered computers. Proc. Int. Conf. VLSI Design (2014). Google ScholarGoogle ScholarDigital LibraryDigital Library
  29. [29] Khanna Sudhanshu et al. 2014. An FRAM-based nonvolatile logic MCU SoC exhibiting 100% digital state retention at VDD=0 V achieving zero leakage with \(\lt\)400-ns wakup time for ULP applications. IEEE J. Solid-State Circuits 49, 1 (Jan. 2014). Google ScholarGoogle ScholarCross RefCross Ref
  30. [30] Liang Ming and Hu Xiaolin. 2015. Recurrent convolutional neural network for object recognition. In Proc. IEEE Conf. Comput. Vis, Pattern Recogn. 33673375.Google ScholarGoogle Scholar
  31. [31] Liu Yongpan, Wang Zhibo, et al. 2016. A 65nm ReRAM-enabled nonvolatile processor with 6\(\times\) reduction in restore time and 4\(\times\) higher clock frequency using adaptive data retention and self-write-termination nonvolatile logic. In Proc. ISSCC Conf.IEEE. Google ScholarGoogle ScholarCross RefCross Ref
  32. [32] Majid Amjad Yousef, Donne Carlo Delle, Maeng Kiwan, Colin Alexei, Yildirim Kasim Sinan, Lucia Brandon, and Pawełczak Przemysław. 2020. Dynamic task-based intermittent execution for energy-harvesting devices. ACM Trans. Sens. Netw. (TOSN) 16, 1 (2020), 124.Google ScholarGoogle ScholarDigital LibraryDigital Library
  33. [33] Meli M. and Bachmann P.. 2016. Powering Long Range Wireless Nodes with Harvested Energy. Technical Report, November. Zürcher Hochschule für Angewandte Wissenschaften (ZHAW).Google ScholarGoogle Scholar
  34. [34] Meyer Matthias et al. 2019. Event-triggered natural hazard monitoring with convolutional neural networks on the edge. In Proc. IPSN Conf. IEEE, 7384.Google ScholarGoogle Scholar
  35. [35] Moser Clemens, Thiele Lothar, Brunelli Davide, and Benini Luca. 2010. Adaptive power management for environmentally powered system. IEEE Trans. Comput. 59, 4 (2010), 478491.Google ScholarGoogle ScholarDigital LibraryDigital Library
  36. [36] Naderiparizi Saman, Parks Aaron N., Kapetanovic Zerina, Ransford Benjamin, and Smith Joshua R.. 2015. WISPCam: A battery-free RFID camera. In Proc. IEEE RFID.Google ScholarGoogle Scholar
  37. [37] OmniVision 2006. OV7670/OV7171 CMSO VGA CameraChip Sensor Preliminary Datasheet. OmniVision. Version 1.4.Google ScholarGoogle Scholar
  38. [38] O’Shea Keiron and Nash Ryan. 2015. An introduction to convolutional neural networks. arXiv:1511.08458.Google ScholarGoogle Scholar
  39. [39] Qazi Masood, Amerasekera Ajith, and Chandrakasan Anantha P.. 2014. A 3.4-pJ FeRAM-enabled D flip-flop in 0.13-\(\mu\)m CMOS for nonvolatile processing in digital systems. IEEE J. Solid-State Circuits 49, 1 (Jan.2014), 202211. Google ScholarGoogle ScholarCross RefCross Ref
  40. [40] Raghunathan Vijay, Kansal Aman, Hsu Jason, Friedman Jonathan, and Srivastava Mani. 2005. Design considerations for solar energy harvesting wireless embedded systems. In Proc. IPSN Conf. IEEE Press, 64.Google ScholarGoogle Scholar
  41. [41] Ransford Benjamin, Sorber Jacob, and Fu Kevin. 2011. Mementos: System support for long-running computation on RFID-scale devices. SIGARCH Comput. Archit. News 46, 3 (March2011), 12. Google ScholarGoogle ScholarDigital LibraryDigital Library
  42. [42] Arreola Alberto Rodriguez, Balsamo Domenico, Merrett Geoff V., and Weddell Alex S.. 2018. RESTOP: Retaining external peripheral state in intermittently-powered sensor systems. Sensors 18, 1 (2018).Google ScholarGoogle Scholar
  43. [43] Rosen Barry K., Wegman Mark N., and Zadeck F. Kenneth. 1988. Global value numbers and redundant computations. In Proc. 15th ACM SIGPLAN-SIGACT Symp. Principles Program. Lang. ACM, 1227.Google ScholarGoogle Scholar
  44. [44] Sigrist Lukas, Ahmed Rehan, Gomez Andres, and Thiele Lothar. 2020. Harvesting-aware optimal communication scheme for infrastructure-less sensing. ACM Trans. Internet Things 1, 4 (2020), 126.Google ScholarGoogle ScholarDigital LibraryDigital Library
  45. [45] Sigrist Lukas, Gomez Andres, Leubin Matthias, Beutel Jan, and Thiele Lothar. 2021. Environment and application testbed for low-power energy harvesting system design. IEEE Trans. Indust. Electron. 68, 11 (2021), 1114611156. Google ScholarGoogle ScholarCross RefCross Ref
  46. [46] Sigrist Lukas, Gomez Andres, Lim Roman, Lippuner Stefan, Leubin Matthias, and Thiele Lothar. 2017. Measurement and validation of energy harvesting IoT devices. In Proc. DATE Conf. IEEE, 11591164.Google ScholarGoogle Scholar
  47. [47] Sigrist Lukas, Gomez Andres, and Thiele Lothar. 2019. Dataset: Tracing indoor solar harvesting. In Proc. 2nd Worksh. Data Acquis. Anal. (DATA’19). ACM. Google ScholarGoogle ScholarDigital LibraryDigital Library
  48. [48] Spies Peter, Pollak Markus, and Mateu Loreto. 2015. Handbook of Energy Harvesting Power Supplies and Applications. CRC Press.Google ScholarGoogle ScholarCross RefCross Ref
  49. [49] Stricker Naomi and Thiele Lothar. 2020. Analysing and improving robustness of predictive energy harvesting systems. In Proc. ENSsys Worksh. Association for Computing Machinery, New York, NY.Google ScholarGoogle Scholar
  50. [50] Tretter Andreas. 2018. On Efficient Data Exchange in Multicore Architectures. Ph.D. Dissertation. ETH Zurich.Google ScholarGoogle Scholar
  51. [51] Verykios Theodoros D., Balsamo Domenico, and Merrett Geoff V.. 2018. Selective policies for efficient state retention in transiently-powered embedded systems: Exploiting properties of NVM technologies. Sustain. Comput. Inform. Syst. 22 (2018), 167–178. Google ScholarGoogle ScholarCross RefCross Ref
  52. [52] Weddell A. S., Magno M., Merrett G. V., Brunelli D., Al-Hashimi B. M., and Benini L.. 2013. A survey of multi-source energy harvesting systems. In Proc. DATE Conf.Google ScholarGoogle ScholarCross RefCross Ref
  53. [53] Wojek Christian, Dorkó Gyuri, Schulz André, and Schiele Bernt. 2008. Sliding-windows for rapid object class localization: A parallel technique. In Joint Pattern Recognition Symposium. Springer, 7181.Google ScholarGoogle Scholar
  54. [54] Wong H-S. Philip, Raoux Simone, Kim SangBum, Liang Jiale, Reifenberg John P., Rajendran Bipin, Asheghi Mehdi, and Goodson Kenneth E.. 2010. Phase change memory. Proc. IEEE 98, 12 (2010), 22012227.Google ScholarGoogle ScholarCross RefCross Ref

Index Terms

  1. Dataflow Driven Partitioning of Machine Learning Applications for Optimal Energy Use in Batteryless Systems

    Recommendations

    Comments

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Sign in

    Full Access

    • Published in

      cover image ACM Transactions on Embedded Computing Systems
      ACM Transactions on Embedded Computing Systems  Volume 21, Issue 5
      September 2022
      526 pages
      ISSN:1539-9087
      EISSN:1558-3465
      DOI:10.1145/3561947
      • Editor:
      • Tulika Mitra
      Issue’s Table of Contents

      Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

      Publisher

      Association for Computing Machinery

      New York, NY, United States

      Publication History

      • Published: 9 December 2022
      • Online AM: 21 March 2022
      • Accepted: 19 February 2022
      • Revised: 16 February 2022
      • Received: 30 July 2021
      Published in tecs Volume 21, Issue 5

      Permissions

      Request permissions about this article.

      Request Permissions

      Check for updates

      Qualifiers

      • research-article
      • Refereed

    PDF Format

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader

    Full Text

    View this article in Full Text.

    View Full Text

    HTML Format

    View this article in HTML Format .

    View HTML Format
    About Cookies On This Site

    We use cookies to ensure that we give you the best experience on our website.

    Learn more

    Got it!