Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{185925,
        author = {Mandeep Kaur},
        title = {Hybrid resource allocation framework using Deep reinforcement learning for cloud data centers},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {5},
        pages = {3733-3743},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=185925},
        abstract = {Cloud data centers are confronted with mounting challenges in dealing with dynamic workloads while being energy-efficient, SLA-compliant, and low-latency. Conventional methods of allocating resources tend to be ineffective in adapting to the demands, resulting in underutilization and performance degradation. To mitigate this, we suggest a Hybrid Resource Allocation Framework that combines Deep Reinforcement Learning with heuristic optimization to improve resource use and system responsiveness. The system designs the allocation process by utilizing a Markov Decision Process (MDP) and learning optimal CPU, memory, and task scheduling policies by the DRL agent while the heuristic layer manages SLA-critical and latency-sensitive requests. Experimental analysis on the Google Cluster Dataset (2019) shows dramatic improvements: 32% greater resource utilization, 28% less SLA violations, 25% power savings, and 41% latency improvement over conventional approaches. Overall, the suggested hybrid framework provides a scalable, adaptive, and energy-efficient approach to next-generation cloud data centers.},
        keywords = {Resource Allocation, Markov Decision Process, Deep Reinforcement Learning, Energy Efficiency},
        month = {October},
        }