Data-Driven Cloud Workload Optimization Using Machine Learning Modeling for Proactive Resource Management
DOI:
https://doi.org/10.63282/3050-922X.IJERET-V6I4P104Keywords:
Cloud Workload Optimization, Proactive Resource Management, Machine Learning, Resource Allocation, Time Series Modeling, Cloud ComputingAbstract
Cloud computing has reinvented delivery of services due to its provision of scalability, elasticity and cost effectiveness. But increased workloads have aggravated management of resources, which has resulted in energy waste, SLA break, and high cost of operation. The study presents an active data-driven optimization model based on machine learning to predict workload dynamics. Google Cluster Trace dataset is processed using logarithmic scaling, Savitzky-Golay filtering and Min-max normalization in order to improve stability and quality. An LSTM model is used to find longitudinal dependencies and forecast workload, CPU, and RAM usage. Measures of evaluation are R2, Mean Squared error (MSE) and Root mean Squared logarithmic error (RMSLE). The findings indicate that LSTM attains almost perfect accuracy, R2 of 0.99%, MSE of 13,934.54 (workload), 128.89 (CPU) and 131.29 (RAM), and RMSLE of 0.15, 0.16 and 0.14. Compared to existing models such as VAMBig, SVM, and SATCN, the LSTM framework significantly outperforms them in prediction accuracy and error minimization. These findings confirm that proactive LSTM-based workload optimization reduces energy usage, enhances SLA compliance, and strengthens scalability in dynamic cloud computing environments. Future work can extend this framework to multi-cloud and federated environments for broader applicability
References
[1] M. G. Avram, “Advantages and Challenges of Adopting Cloud Computing from an Enterprise Perspective,” Procedia Technol., 2014, doi: 10.1016/j.protcy.2013.12.525.
[2] V. Prajapati, “Cloud-Based Database Management: Architecture, Security, challenges and solutions,” J. Glob. Res. Electron. Commun., vol. 01, no. 1, pp. 07–13, 2025.
[3] V. Varma, “Secure Cloud Computing with Machine Learning and Data Analytics for Business Optimization,” ESP J. Eng. Technol. Adv., vol. 4, no. 3, 2024, doi: 10.56472/25832646/JETA-V4I3P119.
[4] M. Pantazoglou, G. Tzortzakis, and A. Delis, “Decentralized and energy-efficient workload management in enterprise clouds,” IEEE Trans. cloud Comput., vol. 4, no. 2, pp. 196–209, 2015.
[5] V. Singh, “Reinventing Business with Cloud Integration: The Cost - Effectiveness of Replacing Legacy Applications,” Int. J. Sci. Res., vol. 13, no. 8, pp. 1882–1887, 2024.
[6] K. Anderson, “Multi-Agent Reinforcement Learning for Enterprise Cloud Workload Optimization,” 2023.
[7] G. Maddali, “An Efficient Bio-Inspired Optimization Framework for Scalable Task Scheduling in Cloud Computing Environments,” Int. J. Curr. Eng. Technol., vol. 15, no. 3, pp. 229–238, 2025.
[8] S. Narang and V. G. Kolla, “Next-Generation Cloud Security: A Review of the Constraints and Strategies in Serverless Computing,” Int. J. Res. Anal. Rev., vol. 12, no. 3, pp. 1–7, 2025, doi: 10.56975/ijrar.v12i3.319048.
[9] V. M. L. G. Nerella, “A Database-Centric CSPM Framework for Securing Mission-Critical Cloud Workloads,” Int. J. Intell. Syst. Appl. Eng., vol. 10, no. 1, pp. 209–217, 2022.
[10] A. R. Duggasani, “Scalable and Optimized Load Balancing in Cloud Systems: Intelligent Nature-Inspired Evolutionary Approach,” Int. J. Innov. Sci. Res. Technol., vol. 10, no. 5, pp. 2153–2160, May 2025, doi: 10.38124/ijisrt/25may1290.
[11] R. Dattangire, R. Vaidya, D. Biradar, and A. Joon, “Exploring the Tangible Impact of Artificial Intelligence and Machine Learning: Bridging the Gap between Hype and Reality,” in 2024 1st International Conference on Advanced Computing and Emerging Technologies (ACET), IEEE, Aug. 2024, pp. 1–6. doi: 10.1109/ACET61898.2024.10730334.
[12] A. Birhade, V. Shejul, D. Chavan, N. Y. Patil, and D. R. D. Jadhav, “AI and Machine Learning in Cloud Optimization.” 2025. doi: 10.2139/ssrn.5321423.
[13] R. Q. Majumder, “Machine Learning for Predictive Analytics: Trends and Future Directions,” Int. J. Innov. Sci. Res. Technol., vol. 10, no. 4, pp. 3557–3564, 2025, doi: 10.38124/ijisrt/25apr1899.
[14] N. Prajapati, “The Role of Machine Learning in Big Data Analytics: Tools, Techniques, and Applications,” ESP J. Eng. Technol. Adv., vol. 5, no. 2, pp. 16–22, 2025, doi: 10.56472/25832646/JETA-V5I2P103.
[15] G. Modalavalasa and H. Kali, “Exploring Big Data Role in Modern Business Strategies: A Survey with Techniques and Tools,” Int. J. Adv. Res. Sci. Commun. Technol., vol. 3, no. 3, pp. 431–441, Jan. 2023, doi: 10.48175/IJARSCT-11900B.
[16] A. Sharma and S. Kabade, “AI-Driven and Cloud-Enabled System for Automated Reconciliation and Regulatory Compliance in Pension Fund Management,” Int. J. All Res. Educ. Sci. Methods, vol. 12, no. 6, pp. 3019–3027, 2024.
[17] Q. Xin, “A Deep Reinforcement Learning Approach to Optimizing Cloud Workload Migration,” Am. J. Interdiscip. Res. Innov., vol. 4, no. 3, pp. 10–15, 2025, doi: 10.54536/ajiri.v4i3.5429.
[18] R. Karthikeyan, S. R. A. Samad, V. Balamurugan, S. Balasubaramanian, and R. Cyriac, “Workload Prediction in Cloud Data Centers Using Complex‐Valued Spatio‐Temporal Graph Convolutional Neural Network Optimized With Gazelle Optimization Algorithm,” Trans. Emerg. Telecommun. Technol., vol. 36, no. 3, Mar. 2025, doi: 10.1002/ett.70078.
[19] C. Diwaker and N. Miglani, “Optimizing Autoencoder for Workload Prediction in Cloud Environment Using Particle Swarm Optimization,” in The International Conference on Recent Innovations in Computing, 2024, pp. 185–205.
[20] T. Ali, H. U. Khan, F. K. Alarfaj, and M. Alreshoodi, “Hybrid deep learning and evolutionary algorithms for accurate cloud workload prediction,” Computing, vol. 106, no. 12, pp. 3905–3944, 2024.
[21] Z. Ahamed, M. Khemakhem, F. Eassa, F. Alsolami, A. Basuhail, and K. Jambi, “Deep Reinforcement Learning for Workload Prediction in Federated Cloud Environments.,” Sensors (Basel)., vol. 23, no. 15, Aug. 2023, doi: 10.3390/s23156911.
[22] J. Dogani, F. Khunjush, M. R. Mahmoudi, and M. Seydali, “Multivariate workload and resource prediction in cloud computing using CNN and GRU by attention mechanism,” J. Supercomput., vol. 79, no. 3, pp. 3437–3470, 2023.
[23] S. Priyadarshini, T. N. Sawant, G. Bhimrao Yadav, J. Premalatha, and S. R. Pawar, “Enhancing security and scalability by AI/ML workload optimization in the cloud,” Cluster Comput., vol. 27, no. 10, pp. 13455–13469, Dec. 2024, doi: 10.1007/s10586-024-04641-x.
[24] P. Rawat, “Workload prediction for cloud services by using a hybrid neural network model,” National College of Ireland, 2022.
[25] S. Simaiya et al., “A hybrid cloud load balancing and host utilization prediction method using deep learning and optimization techniques,” Sci. Rep., vol. 14, no. 1, p. 1337, 2024.
[26] Y. Xing, “Work scheduling in cloud network based on deep Q-LSTM models for efficient resource utilization,” J. Grid Comput., vol. 22, no. 1, p. 36, 2024.
[27] M. E. Karim, M. M. S. Maswood, S. Das, and A. G. Alharbi, “BHyPreC: A Novel Bi-LSTM Based Hybrid Recurrent Neural Network Model to Predict the CPU Workload of Cloud Virtual Machine,” IEEE Access, 2021, doi: 10.1109/ACCESS.2021.3113714.
[28] A. Lopez Garcia et al., “A Cloud-Based Framework for Machine Learning Workloads and Applications,” IEEE Access, vol. 8, pp. 18681–18692, 2020, doi: 10.1109/ACCESS.2020.2964386.
[29] J. Bi, H. Ma, H. Yuan, and J. Zhang, “Accurate Prediction of Workloads and Resources with Multi-Head Attention and Hybrid LSTM for Cloud Data Centers,” IEEE Trans. Sustain. Comput., vol. 8, no. 3, pp. 375–384, 2023, doi: 10.1109/TSUSC.2023.3259522.
[30] Z. Amekraz and M. Y. Hadi, “CANFIS: A Chaos Adaptive Neural Fuzzy Inference System for Workload Prediction in the Cloud,” IEEE Access, vol. 10, pp. 49808–49828, 2022, doi: 10.1109/ACCESS.2022.3174061.
[31] H. Yuan, S. Member, J. Bi, S. Member, S. Li, and S. Member, “An Improved LSTM-Based Prediction Approach for Resources and Workload in Large-Scale Data Centers,” vol. 11, no. 12, pp. 22816–22829, 2024.
