Literature review: Transformers in Autoscaling

This is a literature review of 11 papers on transformer-based cloud autoscaling, with a big-picture comparison and a pedagogical reading order.

Transformers applied to autoscaling¶

A taxonomy, comparison table, and the MAPE-K walk-through has been prepared. The reviewed papers are summarized in the following table:

#	Note	Full title	Venue
1	Predictive Auto-scaler for Kubernetes Shim et al. (2023)	Predictive Auto-scaler for Kubernetes Cloud	IEEE International Systems Conference (SysCon)
2	Predictive Autoscaling with Transformers Shrestha & Tuz Sabiha (2025)	Enhancing Cloud Resource Utilization with Predictive Autoscaling Using Transformer Models	9th Int. Conf. on Cloud, Big Data and Communication Systems (ICCBDCS)
3	InformerAutoScale Kumar et al. (2025)	Optimizing resource allocation in cloud-native applications through proactive autoscaling with the InformerAutoScale model	The Journal of Supercomputing 81(9):1077
4	MV-Transformer MAPE Framework Kumar et al. (2025)	A multivariate transformer-based monitor-analyze-plan-execute (MAPE) autoscaling framework for dynamic resource allocation in cloud environment	Computing 107(3):69
5	Transformer Workload Prediction & Adaptive Auto-Scaling G et al. (2026)	Transformer-Based Workload Prediction and Adaptive Auto-Scaling in Cloud Data Centers	IEEE Int. Conf. on Computing, Communication, Control and Networking (I3CTCON)
6	CATScaler Meng et al. (2025)	CATScaler: A Convolution-Augmented Transformer Scaling Framework for Cloud-Native Applications	IEEE Transactions on Services Computing 18(5):2659–2672
7	WGAN-gp Transformer Arbat et al. (2022)	Wasserstein Adversarial Transformer for Cloud Workload Prediction	Proceedings of the AAAI Conference on Artificial Intelligence 36(11):12433–12439
8	Diffusion Autoformer (job-arrival) Kumar et al. (2025)	Predicting Cloud Workload Job Arrival Rates Using a Diffusion Autoformer Model	IEEE International Conference on Big Data (BigData)
9	Fremer Ye et al. (2025)	Fremer: Lightweight and Effective Frequency Transformer for Workload Forecasting in Cloud Services	Proceedings of the VLDB Endowment 18(11):3812–3825
10	FELT Ding et al. (2025)	FELT: Large-Scale Cloud Workload Prediction Through Adaptive Feature-Enhanced and Similarity-Aware Transformer	Tsinghua Science and Technology
11	CloudFormer Shahbazinia et al. (2025)	CloudFormer: An Attention-based Performance Prediction for Public Clouds with Unknown Workload	Preprint, arXiv:2509.03394

Forecasting foundations¶

A comparison of forecasting models is available. The general time-series forecasting transformers that may underpin the autoscaling forecasters above are:

#	Note	Full title	Venue
1	Autoformer Wu et al. (2021)	Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting	NeurIPS 2021
2	Informer Zhou et al. (2020)	Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting	AAAI 2021 (arXiv:2012.07436)
3	PatchTST Nie et al. (2022)	A Time Series is Worth 64 Words: Long-term Forecasting with Transformers	ICLR 2023 (arXiv:2211.14730)
4	TimesFM Das et al. (2023)	A decoder-only foundation model for time-series forecasting	ICML 2024 (arXiv:2310.10688)

References¶

Shim, S., Dhokariya, A., Doshi, D., Upadhye, S., Patwari, V., & Park, J.-Y. (2023). Predictive Auto-scaler for Kubernetes Cloud. 2023 IEEE International Systems Conference (SysCon), 1–8. 10.1109/syscon53073.2023.10131106
Shrestha, R., & Tuz Sabiha, F. (2025). Enhancing Cloud Resource Utilization with Predictive Autoscaling Using Transformer Models. 2025 9th International Conference on Cloud and Big Data Computing (ICCBDC), 24–29. 10.1109/iccbdc67784.2025.00011
Kumar, S., Chauhan, M. K., Priyadarshni, Tripathi, S., Misra, R., & Singh, T. N. (2025). Predicting Cloud Workload Job Arrival Rates Using a Diffusion Autoformer Model. 2025 IEEE International Conference on Big Data (BigData), 6102–6107. 10.1109/bigdata66926.2025.11401693
G, C. N., O, B. C., R, J. K., Raj B N, P., Naik, P. N., & G, M. B. (2026). Transformer-Based Workload Prediction and Adaptive Auto-Scaling in Cloud Data Centers. 2026 IEEE International Conference for Convergence in Computing Technology (I3CTCON), 1–6. 10.1109/i3ctcon68242.2026.11507247
Meng, F., Dai, H., Cong, G., Zhu, B., & Zhao, H. (2025). CATScaler: A Convolution-Augmented Transformer Scaling Framework for Cloud-Native Applications. IEEE Transactions on Services Computing, 18(5), 2659–2672. 10.1109/tsc.2025.3592383
Arbat, S., Jayakumar, V. K., Lee, J., Wang, W., & Kim, I. K. (2022). Wasserstein Adversarial Transformer for Cloud Workload Prediction. Proceedings of the AAAI Conference on Artificial Intelligence, 36(11), 12433–12439. 10.1609/aaai.v36i11.21509
Ye, H., Chen, J., Jiang, F., He, X., Zhang, T., Chen, J., & Gao, X. (2025). Fremer: Lightweight and Effective Frequency Transformer for Workload Forecasting in Cloud Services. Proceedings of the VLDB Endowment, 18(11), 3812–3825. 10.14778/3749646.3749656
Ding, Z., Feng, B., & Yu, W. (2025). FELT: Large-Scale Cloud Workload Prediction Through Adaptive Feature-Enhanced and Similarity-Aware Transformer. Tsinghua Science and Technology. 10.26599/tst.2025.9010102
Shahbazinia, A., Huang, D., Costero, L., & Atienza, D. (2025). CloudFormer: An Attention-based Performance Prediction for Public Clouds with Unknown Workload. arXiv. 10.48550/ARXIV.2509.03394
Wu, H., Xu, J., Wang, J., & Long, M. (2021). Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting. arXiv. 10.48550/ARXIV.2106.13008
Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H., & Zhang, W. (2020). Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. arXiv. 10.48550/ARXIV.2012.07436
Nie, Y., Nguyen, N. H., Sinthong, P., & Kalagnanam, J. (2022). A Time Series is Worth 64 Words: Long-term Forecasting with Transformers. arXiv. 10.48550/ARXIV.2211.14730
Das, A., Kong, W., Sen, R., & Zhou, Y. (2023). A decoder-only foundation model for time-series forecasting. arXiv. 10.48550/ARXIV.2310.10688