Energy-Aware AI Scheduling for Resource-Constrained Edge Devices
DOI:
https://doi.org/10.63282/3050-922X.ICRCEDA25-110Keywords:
Edge AI, Energy-Aware Scheduling, Resource-Constrained Devices, Real-Time Systems, Task Scheduling, Energy Efficiency, Embedded AI, Edge Computing, Dynamic Workload ManagementAbstract
As artificial intelligence (AI) applications continue to proliferate at the edge of networks, ensuring efficient utilization of limited computational and energy resources has become critical. This paper proposes a novel energy-aware AI scheduling framework tailored for resource-constrained edge devices. Our approach dynamically allocates computational tasks based on energy consumption models, workload characteristics, and system performance constraints. We integrate lightweight profiling techniques with real-time scheduling algorithms to balance energy efficiency and task accuracy. Experimental results on representative edge hardware platforms show that our method reduces energy consumption by up to 35% while maintaining comparable AI performance, outperforming traditional fixed-scheduling approaches. These findings highlight the potential of intelligent scheduling strategies in enabling sustainable and scalable edge AI deployment
References
[1] Y. Kang, J. Hauswald, C. Gao, A. Rovinski, T. Mudge, and J. Mars, "Neurosurgeon: Collaborative intelligence between the cloud and mobile edge," ACM SIGARCH Computer Architecture News, vol. 45, no. 1, pp. 615–629, 2017.
[2] A. Mittal and A. Verma, "Dynamic energy-aware scheduling for real-time systems on DVS-enabled processors," Proceedings of the International Conference on Embedded Software and Systems, pp. 417–426, 2010.
[3] S. Banerjee, K. Bhardwaj, and T. Mitra, "Power-aware deployment and scheduling of embedded deep neural networks," ACM Transactions on Embedded Computing Systems (TECS), vol. 18, no. 5s, pp. 1–23, 2019.
[4] H. Esmaeilzadeh, E. Blem, R. Amant, K. Sankaralingam, and D. Burger, "Dark silicon and the end of multicore scaling," Proceedings of the 38th Annual International Symposium on Computer Architecture (ISCA), pp. 365–376, 2011.
[5] S. Han, H. Mao, and W. J. Dally, "Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding," International Conference on Learning Representations (ICLR), 2016.
[6] Z. Wu, D. An, W. Wu, and X. Li, "An adaptive energy-efficient scheduling mechanism for real-time tasks on multicore embedded systems," IEEE Transactions on Industrial Informatics, vol. 14, no. 3, pp. 1026–1037, 2018.
[7] X. Zhang, Z. Lin, and H. Li, "A survey on energy-efficient scheduling for real-time systems," Journal of Systems Architecture, vol. 90, pp. 71–84, 2018.
[8] Y. Xiao, D. Jin, and Y. Yang, "Edge computing security: State of the art and challenges," Proceedings of the IEEE, vol. 107, no. 8, pp. 1608–1631, 2019.
[9] A. Ignatov et al., "AI benchmark: Running deep neural networks on Android smartphones," European Conference on Computer Vision (ECCV) Workshops, pp. 0–15, 2018.
[10] M. Samragh, J. K. Kim, and F. Koushanfar, "Collaborative privacy-preserving deep learning with unmanned aerial vehicles," Proceedings of the 17th ACM Conference on Embedded Networked Sensor Systems (SenSys), pp. 1–13, 2019.
[11] F. Zhou, Y. Huang, Q. Wu, and W. Yang, "Energy-efficient task offloading and resource allocation for mobile edge computing," IEEE Transactions on Vehicular Technology, vol. 67, no. 11, pp. 11127–11140, 2018.
[12] T. Zhang, S. Ye, Y. Wang, and S. Hu, "Efficient scheduling of deep learning workloads on edge devices," IEEE Internet of Things Journal, vol. 8, no. 6, pp. 4430–4441, 2021.
[13] M. Horowitz, "1.1 Computing's energy problem (and what we can do about it)," IEEE International Solid-State Circuits Conference Digest of Technical Papers (ISSCC), pp. 10–14, 2014.
[14] H. Li, Z. Chen, and H. Zhang, "A survey on energy-efficient computing and resource management in edge AI," ACM Computing Surveys (CSUR), vol. 54, no. 10, pp. 1–36, 2022.
[15] A. Sinha and A. Chandrakasan, "Dynamic power management in wireless sensor networks," IEEE Design & Test of Computers, vol. 18, no. 2, pp. 62–74, 2001.
[16] Animesh Kumar, “Redefining Finance: The Influence of Artificial Intelligence (AI) and Machine Learning (ML)”, Transactions on Engineering and Computing Sciences, 12(4), 59-69. 2024.
[17] Vegineni, Gopi Chand, and Bhagath Chandra Chowdari Marella. "Integrating AI-Powered Dashboards in State Government Programs for Real-Time Decision Support." AI-Enabled Sustainable Innovations in Education and Business, edited by Ali Sorayyaei Azar, et al., IGI Global, 2025, pp. 251-276. https://doi.org/10.4018/979-8-3373-3952-8.ch011
[18] Palakurti, A., & Kodi, D. (2025). “Building intelligent systems with Python: An AI and ML journey for social good”. In Advancing social equity through accessible green innovation (pp. 1–16). IGI Global.
[19] Mohanarajesh, Kommineni (2024). Generative Models with Privacy Guarantees: Enhancing Data Utility while Minimizing Risk of Sensitive Data Exposure. International Journal of Intelligent Systems and Applications in Engineering 12 (23):1036-1044.
[20] Puvvada, R. K. (2025). Enterprise Revenue Analytics and Reporting in SAP S/4HANA Cloud. European Journal of Science, Innovation and Technology, 5(3), 25-40.
[21] B. C. C. Marella, “Data Synergy: Architecting Solutions for Growth and Innovation,” International Journal of Innovative Research in Computer and Communication Engineering, vol. 11, no. 9, pp. 10551–10560, Sep. 2023.
[22] A Novel AI-Blockchain-Edge Framework for Fast and Secure Transient Stability Assessment in Smart Grids, Sree Lakshmi Vineetha Bitragunta, International Journal for Multidisciplinary Research (IJFMR), Volume 6, Issue 6, November-December 2024, PP-1-11.
[23] Padmaja Pulivarthy. (2024/12/3). Harnessing Serverless Computing for Agile Cloud Application Development,” FMDB Transactionson Sustainable Computing Systems. 2,( 4), 201-210, FMDB.
[24] Marella, B.C.C., & Kodi, D. (2025). “Fraud Resilience: Innovating Enterprise Models for Risk Mitigation”. Journal of Information Systems Engineering and Management, 10(12s), 683–695.
[25] Mohanarajesh Kommineni, Swathi Chundru, Praveen Kumar Maroju, P Selvakumar. (2025).
Ethical Implications of AI in Sustainable Development Pedagogy, Rethinking the Pedagogy of Sustainable Development in the AI Era, 17-36, IGI Global Scientific Publishing.
[26] Aragani, Venu Madhav and Maroju, Praveen Kumar and Mudunuri, Lakshmi Narasimha Raju, Efficient Distributed Training through Gradient Compression with Sparsification and Quantization Techniques (September 29, 2021). Available at SSRN: https://ssrn.com/abstract=5022841 or http://dx.doi.org/10.2139/ssrn.5022841
[27] Pulivarthy, P. (2024). Optimizing Large Scale Distributed Data Systems Using Intelligent Load Balancing Algorithms. AVE Trends in Intelligent Computing Systems, 1(4), 219–230.
[28] Muniraju Hullurappa, Sudheer Panyaram, “Quantum Computing for Equitable Green Innovation Unlocking Sustainable Solutions,” in Advancing Social Equity Through Accessible Green Innovation, IGI Global, USA, pp. 387- 402, 2025.
[29] MRM Reethu, LNR Mudunuri, S Banala,(2024) "Exploring the Big Five Personality Traits of Employees in Corporates," in FMDB Transactions on Sustainable Management Letters 2 (1), 1-13
[30] Aragani, Venu Madhav and Maroju, Praveen Kumar and Mudunuri, Lakshmi Narasimha Raju, “Efficient Distributed Training through Gradient Compression with Sparsification and Quantization Techniques” (September 29, 2021). Available at SSRN: https://ssrn.com/abstract=5022841 or http://dx.doi.org/10.2139/ssrn.5022841.
[31] Puvvada, R. K. "Optimizing Financial Data Integrity with SAP BTP: The Future of Cloud-Based Financial Solutions." European Journal of Computer Science and Information Technology 13.31 (2025): 101-123.
[32] INNOVATIVE DESIGN OF REFINING MUSCULAR INTERFACES FOR IMPLANTABLE POWER SYSTEMS, Sree Lakshmi Vineetha Bitragunta ,International Journal of Core Engineering & Management, Volume-6, Issue-12, 2021,PP-436-445.
[33] Kirti Vasdev. (2019). “GIS in Disaster Management: Real-Time Mapping and Risk Assessment”. International Journal on Science and Technology, 10(1), 1–8. https://doi.org/10.5281/zenodo.14288561
