AI-Powered Workload Balancing in Vehicular Edge-Cloud Pipelines
DOI:
https://doi.org/10.15680/IJCTECE.2024.0701004Keywords:
Vehicular edge computing, Cloud computing, Workload balancing, Artificial intelligence, Machine learning, Reinforcement learning, Autonomous vehicles, Resource allocation, Real-time systems, Vehicular networksAbstract
The rapid growth of connected and autonomous vehicles demands efficient computing frameworks to handle intensive data processing and real-time decision-making. Vehicular edge-cloud pipelines combine the low latency of edge computing with the high computational capacity of cloud resources to meet these requirements. However, workload distribution across heterogeneous edge and cloud nodes presents challenges in maintaining optimal system performance, reducing latency, and ensuring resource utilization.
This paper proposes an AI-powered workload balancing framework designed specifically for vehicular edge-cloud pipelines. The framework leverages machine learning algorithms to dynamically allocate computing tasks between edge nodes (such as in-vehicle processors and roadside units) and cloud servers based on real-time network conditions, resource availability, and workload characteristics. By analyzing telemetry data, network bandwidth, and task priorities, the AI model predicts optimal distribution strategies that improve throughput and minimize latency.
We design a hybrid workload management system that incorporates reinforcement learning for adaptive decision-making, enabling the pipeline to self-optimize under varying vehicular traffic and network scenarios. The framework supports diverse vehicular applications including real-time perception, data analytics, and infotainment services.
Experimental evaluation using a simulated vehicular network demonstrates that our AI-powered balancing approach outperforms static and heuristic baselines by reducing average task completion time by up to 30%, and improving resource utilization by 25%. Additionally, the system maintains low latency critical for safety-critical autonomous driving functions.
The proposed workload balancing framework advances the state-of-the-art in vehicular edge-cloud orchestration, enabling scalable, efficient, and resilient autonomous vehicle ecosystems. This work lays the foundation for future intelligent vehicular computing platforms that can dynamically adapt to complex operational environments.
References
1. Chen, L., Zhang, Z., & Li, Y. (2021). AI-Driven Resource Allocation for Vehicular Edge-Cloud Computing. IEEE Transactions on Vehicular Technology, 70(4), 3456-3467.
2. Lekkala, C. (2019). Optimizing Data Reliability and Consistency in Hadoop Environments by Introducing ACID Capabilities. European Journal of Advances in Engineering and Technology, 6(5), 73-78.
3. Wu, J., Wang, X., & Liu, Q. (2020). Reinforcement Learning-Based Task Offloading for Mobile Edge Computing. IEEE Transactions on Mobile Computing, 19(6), 1302-1315.
4. Kumbum, P. K., Adari, V. K., Chunduru, V. K., Gonepally, S., & Amuda, K. K. (2023). Navigating digital privacy and security effects on student financial behavior, academic performance, and well-being. Data Analytics and Artificial Intelligence, 3(2), 235–246.
5. Badmus, A., & Adebayo, M. (2020). Compliance-Aware Devops for Generative AI: Integrating Legal Risk Management, Data Controls, and Model Governance to Mitigate Deepfake and Data Privacy Risks in Synthetic Media Deployment.
6. Xu, Y., Sun, S., & Li, M. (2019). Load-Aware Dynamic Offloading in Vehicular Edge Networks. IEEE Access, 7, 127001-127013.
7. Sethupathy, U. K. A. REAL-TIME SUPPLY CHAIN PROCESS AUTOMATION AND MONITORING WITH STREAM PROCESSING.
8. Praveen Kumar, K., Adari, Vijay Kumar., Vinay Kumar, Ch., Srinivas, G., & Kishor Kumar, A. (2024). Optimizing network function virtualization: A comprehensive performance analysis of hardware-accelerated solutions. SOJ Materials Science and Engineering, 10(1), 1-10.-
9. Yang, H., Zhou, Y., & Lin, K. (2018). Heuristic Offloading Strategy for Edge Computing in Connected Vehicles. IEEE Communications Letters, 22(12), 2452-2455.
10. Zhang, Q., & Wang, J. (2022). Hybrid AI Workload Balancing for Autonomous Vehicle Networks. IEEE Internet of Things Journal, 9(3), 2345-2358.
