Issue |
EPJ Web Conf.
Volume 326, 2025
International Conference on Functional Materials and Renewable Energies: COFMER’05 5th Edition
|
|
---|---|---|
Article Number | 05007 | |
Number of page(s) | 4 | |
Section | Smart Energy systems: Storage, Management, Integration | |
DOI | https://doi.org/10.1051/epjconf/202532605007 | |
Published online | 21 May 2025 |
https://doi.org/10.1051/epjconf/202532605007
Hybrid Recurrent Neural Network and Decision Tree Scheduling for Energy-Efficient Resource Allocation in Cloud Computing
1 POLITEHNICA Bucharest, Research Center CAMPUS, 060042 Bucharest, Romania
2 POLITEHNICA Bucharest, Telecommunication Department, 060042 Bucharest, Romania
* Corresponding author: sefati.seyedsalar@upb.ro
Published online: 21 May 2025
Efficient resource allocation in cloud computing is critical for optimizing execution time, minimizing delays, and improving system reliability. Traditional heuristic-based scheduling approaches struggle to adapt to dynamic workloads and heterogeneous virtual machines (VMs), leading to suboptimal performance. This paper proposes a hybrid scheduling framework that integrates Recurrent Neural Networks (RNNs) for execution time prediction and Decision Trees (DTs) for VM classification, enhancing resource allocation efficiency. The RNN model uses historical execution data to accurately predict task execution time, while the DT model classifies VMs based on performance characteristics, ensuring optimal task-to- VM assignments. The proposed method dynamically adapts to workload variations, reducing execution delays and improving Quality of Service (QoS) metrics. Experimental evaluations demonstrate that the hybrid RNN-DT approach outperforms traditional scheduling methods and metaheuristic algorithms, such as Genetic Algorithm and Artificial Bee Colony, in terms of execution time reduction, reliability, and delay minimization.
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.