Energy Harvesting-Supported Efficient Low-Power ML Processing with Adaptive Checkpointing and Intermittent Computing
Pages 1 - 6
Abstract
The rise of ultra-low-power embedded processors has led to increased use of energy harvesting devices (EHDs), providing portability and extended lifespans, but also presenting challenges due to sporadic ambient energy and limited storage. This paper introduces "Micro-Controller Unit - Early Exit Neural Network" MCU-EENet, a framework utilizing kinetic energy harvesting to support machine learning tasks intermittently with early exits. The intermittent nature of ambient energy can lead to potential program interruptions, necessitating efficient state retention techniques within MCU-EENet. We propose "SmartCheck," a memory-optimized runtime checkpointing technique integrated with MCU-EENet to manage and utilize harvested energy efficiently. Through extensive experimentation, our framework demonstrates significant energy savings and performance enhancements, making it a promising solution for energy-constrained environments. Experimental results show a ~ 30% to 50% reduction in energy footprint and a 1.2X to 2.4X speed improvement over existing checkpointing methods.
References
[1]
Domenico Balsamo, Alex S. Weddell, Anup Das, Alberto Rodríguez Arreola, Davide Brunelli, Bashir M. Al-Hashimi, Geoff V. Merrett, and Luca Benini. 2016. Hibernus++: A Self-Calibrating and Adaptive System for Transiently-Powered Embedded Devices. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (2016).
[2]
Domenico Balsamo, Alex S. Weddell, Geoff V. Merrett, Bashir M. Al-Hashimi, Davide Brunelli, and Luca Benini. 2015. Hibernus: Sustaining Computation During Intermittent Supply for Energy-Harvesting Systems. IEEE Embedded Systems Letters (2015).
[3]
Alexei Colin and Brandon Lucia. 2016. Chain: tasks and channels for reliable intermittent programs. ACM SIGPLAN International Conference on Object-Oriented Programming, Systems, Languages, and Applications (2016).
[4]
Abhijitt Dhavlle, Sanket Shukla, Setareh Rafatirad, Houman Homayoun, and Sai Manoj Pudukotai Dinakarrao. 2021. HMD-Hardener: Adversarially Robust and Efficient Hardware-Assisted Runtime Malware Detection. In Design, Automation Test in Europe Conference Exhibition (DATE).
[5]
Zahra Ghodsi, Siddharth Garg, and Ramesh Karri. 2017. Optimal checkpointing for secure intermittently-powered IoT devices. (2017).
[6]
Rosilah Hassan, Faizan Qamar, Mohammad Kamrul Hasan, Azana Hafizah Mohd Aman, and Amjed Sid Ahmed. 2020. Internet of Things and Its Applications: A Comprehensive Survey. Symmetry (2020).
[7]
Matthew Hicks. 2017. Clank: Architectural Support for Intermittent Computation. (2017).
[8]
Andrew G. Howard, Mark Sandler, Grace Chu, Liang-Chieh Chen, Bo Chen, Mingxing Tan, Weijun Wang, Yukun Zhu, Ruoming Pang, Vijay Vasudevan, Quoc V. Le, and Hartwig Adam. 2019. Searching for MobileNetV3. International Conference on Computer Vision (ICCV) (2019).
[9]
Hrishikesh Jayakumar, Arnab Raha, Jacob R. Stevens, and Vijay Raghunathan. 2017. Energy-Aware Memory Mapping for Hybrid FRAM-SRAM MCUs in Intermittently-Powered IoT Devices. (2017).
[10]
Sreenitha Kasarapu, Sanket Shukla, Rakibul Hassan, Avesta Sasan, Houman Homayoun, and Sai Manoj PD. 2022. CAD-FSL: Code-Aware Data Generation based Few-Shot Learning for Efficient Malware Detection. In Great Lakes Symposium on VLSI.
[11]
Seulki Lee, Bashima Islam, Yubo Luo, and Shahriar Nirjon. 2019. Intermittent learning: On-device machine learning on intermittently powered system. Interactive, Mobile, Wearable and Ubiquitous Technologies (2019).
[12]
Ji Lin, Wei-Ming Chen, Han Cai, Chuang Gan, and Song Han. 2021. MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learning. (2021).
[13]
Jie Lin, Wei Yu, Nan Zhang, Xinyu Yang, Hanlin Zhang, and Wei Zhao. 2017. A Survey on Internet of Things: Architecture, Enabling Technologies, Security and Privacy, and Applications. IEEE Internet of Things Journal (2017).
[14]
Brandon Lucia, Vignesh Balaji, Alexei Colin, Kiwan Maeng, and Emily Ruppel. 2017. Intermittent computing: Challenges and opportunities. Summit on Advances in Programming Languages (2017).
[15]
Brandon Lucia and Benjamin Ransford. 2015. A simpler, safer programming and execution model for intermittent systems. ACM SIGPLAN Conference on Programming Language Design and Implementation (2015).
[16]
Benjamin Ransford, Jacob M. Sorber, and Kevin Fu. 2011. Mementos: system support for long-running computation on RFID-scale devices. (2011).
[17]
Muhammad Moid Sandhu, Kai Geissdoerfer, Sara Khalifa, Raja Jurdak, Marius Portmann, and Brano Kusy. 2020. Towards optimal kinetic energy harvesting for the batteryless IoT. (2020).
[18]
Muhammad Shafique, Rehan Hafiz, Muhammad Usama Javed, Sarmad Abbas, Lukas Sekanina, Zdenek Vasicek, and Vojtech Mrazek. 2017. Adaptive and energy-efficient architectures for machine learning: Challenges, opportunities, and research roadmap. (2017).
[19]
Sanket Shukla and Sai Manoj Pudukottai Dinakarrao. 2024. Bring it On: Kinetic Energy Harvesting to Spark Machine Learning Computations in IoTs. In International Symposium on Quality Electronic Design (ISQED).
[20]
Sanket Shukla, Setareh Rafatirad, Houman Homayoun, and Sai Manoj Pudukottai Dinakarrao. 2023. Federated Learning with Heterogeneous Models for On-device Malware Detection in IoT Networks. (2023).
[21]
Priyanka Singla, Shubhankar Suman Singh, and Smruti R. Sarangi. 2019. Flexi-Check: An Adaptive Checkpointing Architecture for Energy Harvesting Devices, (DATE). (2019).
[22]
Mingxing Tan and Quoc V. Le. 2019. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. ArXiv (2019).
[23]
Surat Teerapittayanon, Bradley McDanel, and Hsiang-Tsung Kung. 2016. Branchynet: Fast inference via early exiting from deep neural networks. (2016).
[24]
Joel Van Der Woude and Matthew Hicks. 2016. Intermittent Computation without Hardware Support or Programmer Intervention. (2016).
[25]
Theodoros D. Verykios, Domenico Balsamo, and Geoff V. Merrett. 2019. Selective policies for efficient state retention in transiently-powered embedded systems: Exploiting properties of NVM technologies. Sustainable Computing: Informatics and Systems (2019).
[26]
Maciej Wołczyk, Bartosz Wójcik, Klaudia Bałazy, Igor T Podolak, Jacek Tabor, Marek Śmieja, and Tomasz Trzcinski. 2021. Zero time waste: Recycling predictions in early exit neural networks. Neural Information Processing Systems (2021).
[27]
Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin, and Jian Sun. 2017. ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices. Conference on Computer Vision and Pattern Recognition (2017).
Index Terms
- Energy Harvesting-Supported Efficient Low-Power ML Processing with Adaptive Checkpointing and Intermittent Computing
Index terms have been assigned to the content through auto-classification.
Recommendations
Making sense of intermittent energy harvesting
ENSsys '18: Proceedings of the 6th International Workshop on Energy Harvesting & Energy-Neutral Sensing SystemsBatteryless, energy harvesting sensing devices enable new applications and deployment scenarios with their promise of zero maintenance, long lifetime, and small size. These devices fail often and for variable lengths of time because of the ...
Networking Low-Power Energy Harvesting Devices: Measurements and Algorithms
Recent advances in energy harvesting materials and ultra-low-power communications will soon enable the realization of networks composed of energy harvesting devices. These devices will operate using very low ambient energy, such as energy harvested from ...
Energy Harvesting-assisted Ultra-Low-Power Processing-in-Memory Accelerator for ML Applications
GLSVLSI '24: Proceedings of the Great Lakes Symposium on VLSI 2024The proliferation of Internet of Things (IoT) and edge computing devices has become an essential aspect of our daily routines. Particularly, the rise of wearable technology like smartwatches, health trackers, and smart glasses has contributed ...
Comments
Information & Contributors
Information
Published In
August 2024
384 pages
ISBN:9798400706882
DOI:10.1145/3665314
- Chair:
- Pascal Meinerzhagen,
- Program Chair:
- Kapil Dev,
- Program Co-chair:
- Jerald Yoo
This work is licensed under a Creative Commons Attribution International 4.0 License.
Sponsors
- SIGDA: ACM Special Interest Group on Design Automation
- IEEE CAS
- IEEE EDA
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 09 September 2024
Check for updates
Author Tags
Qualifiers
- Research-article
Conference
ISLPED '24
Sponsor:
ISLPED '24: 29th ACM/IEEE International Symposium on Low Power Electronics and Design
August 5 - 7, 2024
CA, Newport Beach, USA
Acceptance Rates
Overall Acceptance Rate 398 of 1,159 submissions, 34%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 198Total Downloads
- Downloads (Last 12 months)198
- Downloads (Last 6 weeks)198
Reflects downloads up to 06 Oct 2024
Other Metrics
Citations
View Options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in