Citation :

Krishnamoorthy V, Akshaya V, Sivanantham S, Ganagavalli K, "Cost-Efficient Predictive Auto-Scaling Using Transformer-LSTM Fusion Tuned with Bayesian Optimization," International Journal of Engineering Trends and Technology (IJETT), vol. 74, no. 8, pp. 210-221, 2026. Crossref, https://doi.org/10.14445/22315381/IJETT-V74I8P114

Abstract

Cloud applications experience frequent and sometimes unpredictable shifts in demand due to user activity, daily usage cycles, and sudden workload spikes. Traditional autoscaling mechanisms used in cloud environments mostly follow reactive, threshold-based rules. They trigger scaling only after resources begin to saturate, which often leads to slow provisioning, increased response times, and occasional SLA violations during high-traffic periods. When demand drops, these systems may still allocate more resources than necessary, resulting in avoidable costs. To address these issues, this work introduces a predictive autoscaling approach built using a combined Transformer-LSTM model. The fusion model is designed to capture both long-term workload trends and short-term sequential patterns, giving it a more accurate view of workload behavior. A cost optimization function is added to translate multi-step workload predictions into suitable virtual machine allocation decisions while considering cloud pricing and SLA constraints. Bayesian optimization is then used to fine-tune the forecasting model and COF parameters, ensuring an effective balance between accuracy, SLA compliance, and cost savings. Experiments using Google Cluster Trace data show encouraging improvements: forecasting accuracy increases by 18-25%, SLA violations drop by 40-55%, and total costs reduce by 22-35% compared to traditional reactive and single-model predictive methods.

Keywords

Bayesian optimization, Cloud computing, Fusion model, Predictive autoscaling, Resource management, Workload forecasting.

References

[1] Shiva Kumar Chinnam, and Ravindra Karanam, “AI-Driven Predictive Autoscaling in Kubernetes: Reinforcement Learning for Proactive Resource Optimization in Cloud-Native Environments,” International Journal of Scientific Research in Computer Science, Engineering and Information Technology, vol. 8, no. 3, pp. 574-582, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[2] Shivam Kumar Choudhary et al., “Autoscaling Enabled Intelligent Load Balancing in Cloud Computing,” International Journal for Research in Applied Science and Engineering Technology, vol. 13, no. 4, pp. 3979-3984, 2025.
[CrossRef]

[3] Yisel Garí et al., “Reinforcement Learning-based Application Autoscaling in the Cloud: A Survey,” Engineering Applications of Artificial Intelligence, vol. 102, pp. 1-23, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[4] Nilabja Roy, Abhishek Dubey, and Aniruddha Gokhale, “Efficient Autoscaling in the Cloud using Predictive Models for Workload Forecasting,” 2011 IEEE 4^th International Conference on Cloud Computing, Washington, DC, USA, pp. 500-507, 2011.
[CrossRef] [Google Scholar] [Publisher Link]

[5] Saleha Alharthi et al., “Auto-Scaling Techniques in Cloud Computing: Issues and Research Directions,” Sensors, vol. 24, no. 17, pp. 1-21, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[6] Dishant Padalia, Abhishek Mazumdar, Bharati Singh, “A CNN-LSTM Combination Network for Cataract Detection using Eye Fundus Images,” arXiv preprint, pp. 1-8, 2022.
[CrossRef] [Google Scholar] [Publisher Link] 

[7] Yidong Chai et al., “Glaucoma Diagnosis in the Chinese Context: An Uncertainty Information-Centric Bayesian Deep Learning Model,” Information Processing and Management, vol. 58, no. 2, 2021.
[CrossRef] [Google Scholar] [Publisher Link]

[8] Yuxing Zhang et al., “Predictive Auto Scaling and Cost Optimization using Machine Learning in AWS Cloud Environments,” Proceedings of the 2^nd International Symposium on Integrated Circuit Design and Integrated Systems, Association for Computing Machinery, New York, NY, United States, pp. 161-167, 2025.
[CrossRef] [Google Scholar] [Publisher Link] 

[9] Narek Badjajian, and Sandy Montajab Hazzouri, “An Effective Workload Prediction with Rnn-Lstm for Efficient Resource Autoscaling in Private Cloud Environments,” International Journal of Advances in Applied Computational Intelligence, vol. 7, no. 1, pp. 63-77, 2025.
[CrossRef] [Google Scholar] [Publisher Link] 

[10] Shivan Singh et al., “ILP Optimized LSTM-based Autoscaling and Scheduling of Containers in Edge-Cloud Environment,” Journal of Telecommunications and Information Technology, no. 2, pp. 56-68, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[11] Yonghua Zhu et al., “A Novel Approach to Workload Prediction using Attention-based LSTM Encoder-Decoder Network in Cloud Environment,” EURASIP Journal on Wireless Communications and Networking, vol. 2019, no. 1, pp. 1-18, 2019.
[CrossRef] [Google Scholar] [Publisher Link]

[12] Ali Yadavar Nikravesh, Samuel A. Ajila, and Chung-Horng Lung, “An Autonomic Prediction Suite for Cloud Resource Provisioning,” Journal of Cloud Computing, vol. 6, no. 1, pp. 1-20, 2017.
[CrossRef] [Google Scholar] [Publisher Link]

[13] Raghavendra Sridhar, Rashi Nimesh Kumar Dhenia, and Ishva Jitendrakumar Kanan, “A Machine Learning Framework for Predictive Workload Modeling and Dynamic Cloud Resource Allocation,” International Journal of Artificial Intelligence, Data Science, and Machine Learning, vol. 4, no. 1, pp. 60-65, 2023.
[CrossRef] [Google Scholar] [Publisher Link]

[14] Jyoti Bawa, Kuljit Kaur Chahal, and Kamaljit Kaur, “Improving Cloud Resource Management: An Ensemble Learning Approach for Workload Prediction,” The Journal of Supercomputing, vol. 81, no. 10, pp. 1-54, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[15] Xin Li et al., “Investigation into Auto-Scaling Mechanisms in Cloud Computing,” Knowledge Science, Engineering and Management: 18^th International Conference, KSEM, Macao, China, pp. 198-209, 2026.
[CrossRef] [Google Scholar] [Publisher Link]

[16] Thang Le Duc, Chanh Nguyen, and Per-Olov Östberg, “Workload Prediction for Proactive Resource Allocation in Large-Scale Cloud-Edge Applications,” Electronics, vol. 14, no. 16, pp. 1-36, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[17] Linyu Zhu et al., “Elastic EDA: Auto-Scaling Cloud Resources for EDA Tasks via Learning-based Approaches,” 2024 IEEE 42^nd International Conference on Computer Design, Milan, Italy, pp. 144-153, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[18] Satya Nagamani Pothu, and Swathi Kailasam, “Hybrid Workload Prediction for Improved Autoscaling in IaaS Clouds: An ARIMA-OLSTM Approach,” Information Systems Engineering, vol. 30, no. 4, pp. 961-970, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[19] Gautam Solaimalai, “AI-Driven Enterprise Architecture for Scalable Cloud Applications,” 2025 International Conference on Recent Innovation in Science Engineering and Technology, Chennai, India, pp. 1-8, 2025.
[CrossRef] [Google Scholar] [Publisher Link]

[20] Siqiao Xue et al., “A Meta Reinforcement Learning Approach for Predictive Autoscaling in the Cloud,” Proceedings of the 28^th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, New York, NY, United States, pp. 4290-4299, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[21] Nhat-Minh Dang-Quang, and Myungsik Yoo, “An Efficient Multivariate Autoscaling Framework using Bi-LSTM for Cloud Computing,” Applied Sciences, vol. 12, no. 7, pp. 1-21, 2022.
[CrossRef] [Google Scholar] [Publisher Link]

[22] István Pintye, József Kovács, and Róbert Lovas, “Enhancing Machine Learning-based Autoscaling for Cloud Resource Orchestration,” Journal of Grid Computing, vol. 22, no. 4, pp. 1-31, 2024.
[CrossRef] [Google Scholar] [Publisher Link]

[23] Mustafa M. Al-Sayed, “Workload Time Series Cumulative Prediction Mechanism for Cloud Resources using Neural Machine Translation Technique,” Journal of Grid Computing, vol. 20, no. 2, pp. 1-29, 2022.
[CrossRef] [Google Scholar] [Publisher Link]