Deep Reinforcement Learning Based Scheduling Framework for Resource Optimization in Heterogeneous Cloud Computing Systems

V. Mukesh

Authors

V. Mukesh Data Analyst, India. Author

Keywords:

Cloud Computing, Deep Reinforcement Learning, Resource Scheduling, Heterogeneous Systems, Resource Optimization, Dynamic Allocation, AI Scheduling Frameworks

Abstract

The rapid expansion of cloud computing has introduced complex scheduling challenges, especially in heterogeneous environments where computational resources vary significantly. Traditional and heuristic-based schedulers often fall short in adapting to dynamic workloads and resource availability. This paper proposes a Deep Reinforcement Learning (DRL) based scheduling framework that learns optimal allocation policies over time to improve resource utilization, minimize latency, and enhance energy efficiency. Simulations on varied workloads show that DRL outperforms conventional algorithms in scalability and responsiveness.

References

1. Mao, H., Alizadeh, M., Menache, I., & Kandula, S. (2016). Resource Management with Deep Reinforcement Learning. Proceedings of the ACM Workshop on Hot Topics in Networks.

2. S. B. Vinay, Natural Language Processing for Legal Documentation in Indian Languages, International Journal of Natural Language Processing (IJNLP), 2(1), 2024, 1-10.

3. Dr. S. Balasubramanian, Facial Recognition Using Artificial Intelligence (AI)- Critical Analysis an Review, International Journal of Graphics and Multimedia (IJGM) 10(1), 2023, pp. 1-6. https://doi.org/10.17605/OSF.IO/7MXU8

4. Li, Q., Wang, J., & Yang, X. (2020). Energy-aware resource allocation in cloud computing using Ant Colony Optimization. Journal of Cloud Computing, 9(1), 1–15.

5. Wang, R., & Singh, A. (2019). Genetic Algorithm-Based Multi-Objective Task Scheduling in Heterogeneous Environments. International Journal of Computer Applications, 182(25).

6. S. B. Vinay, "AI and machine learning integration with AWS SageMaker: current trends and future prospects", International Journal of Artificial Intelligence Tools (IJAIT), vol. 1, issue 1, pp. 1-24, 2024.

7. Chen, H., Hu, Y., & Liu, J. (2021). Hierarchical Deep Reinforcement Learning for Multi-Level Cloud Scheduling. IEEE Transactions on Parallel and Distributed Systems, 32(10), 2350–2362.

8. Manivannan, S., & Balasubramanian, S. (2010). Defect / Effort Prediction Models in Software Maintenance Projects. International Journal of Computer Engineering & Technology (IJCET), 1(1), 20–34.

9. Vinay, S. B. (2024). AI-Driven Patent Mining: Unveiling Innovation Patterns through Automated Knowledge Extraction. International Journal of Super AI (IJSAI), 1(1), 111.

10. Liu, Y., Zhao, W., & Tang, X. (2018). A hybrid scheduling method using DRL and heuristic for cloud computing. Future Generation Computer Systems, 86, 310–320.

11. Zhang, K., & Lee, K. (2017). Dynamic Cloud Workload Management Using Reinforcement Learning. International Conference on High Performance Computing.

12. Xu, Y., & Wei, G. (2019). Reinforcement Learning in Fog and Edge Computing. IEEE Access, 7, 129665–129678.

13. Manivannan, S., & Balasubramanian, S. (2010). Software Process and Product Quality Assurance in Software Organizations. International Journal of Information Technology and Management Information System (IJITMIS), 1(1), 1–11.

14. Kumar, S., & Basu, A. (2020). DRL-based task offloading in mobile edge computing. Computer Communications, 149, 246–258.

15. S.B. Vinay, "Data Scientist Competencies and Skill Assessment: A Comprehensive Framework," International Journal of Data Scientist (IJDST), vol. 1, issue 1, pp. 1-11, 2024.

16. He, L., Chen, T., & Zhou, Y. (2021). A deep Q-learning based scheduler for adaptive resource management in edge-cloud environments. Sensors, 21(4), 1342.

17. Pan, Y., & Wu, J. (2022). Federated Deep Reinforcement Learning for Edge Computing. IEEE Transactions on Mobile Computing, 21(7), 2511–2523.

18. Kabilan, R.(2025). Harnessing Elastic Resource Allocation in Cloud Computing for Scalable Real-Time Analytics in Distributed Systems. Global Journal of Multidisciplinary Research and Development, 6(3), 49–53

19. Singh, R., & Gupta, A. (2020). Adaptive cloud workload distribution using DRL. International Journal of Information Management, 53, 102142.

20. Pradip Kumar Krishnadevarajan, Deepak Muthukrishnan, S. Balasubramanian and N. Kannan. Supply Chain in India “2011-2015” – A Review: Challenges, Solution Framework and Key Best Practices, International Journal of Management, 6(10), 2015, pp. 135-149.

21. Lin, Y., & Wang, C. (2018). Reward Shaping for Task Scheduling Using Reinforcement Learning. Journal of Systems Architecture, 91, 54–62.

22. Nair, V., & Sharma, M. (2021). Multi-agent reinforcement learning in multi-cloud systems. Applied Soft Computing, 107, 107404.

23. Das, A., & Mohapatra, S. (2019). Comparative study of heuristic and DRL in real-time scheduling. International Journal of Artificial Intelligence Tools, 28(3).

24. S. B. Vinay, Application of Artificial Intelligence (AI) In Publishing Industry in India, International Journal of Computer Engineering and Technology (IJCET) 14(1), 2023, pp. 7-12.DOI: https://doi.org/10.17605/OSF.IO/4D5M7

25. Balasubramanian, S., & Francis, J. J. (2011). Impact of Technology on Productivity and Service Quality Among Indian Airline Services. International Journal of Management (IJM), 2(2), 33–43.

26. Gao, M., & Yu, Z. (2023). Transfer learning approaches in DRL for cloud systems. ACM Transactions on Internet Technology, 23(1), 1–23.