AI-Driven Storage Optimization for SAP Workloads in Multi-Cloud Environments
Keywords:
Cloud Erp, Machine Learning, Storage Optimization, Multi-Cloud, Sap S/4hana, Data Tiering, Reinforcement Learning, LstmAbstract
Enterprise Resource Planning (ERP) platforms such as SAP S/4HANA generate high-volume, heterogeneous workloads that impose substantial demands on storage infrastructure in cloud environments. Conventional storage management approaches rely on static tiering and rule-based data lifecycle policies, which are fundamentally limited in their ability to adapt to the dynamic and non-uniform access patterns characteristic of large-scale ERP deployments. These limitations result in measurable inefficiencies in cost, latency, and overall resource utilization. This paper proposes an artificial intelligence (AI)-driven storage optimization framework that employs machine learning (ML) models to predict workload behavior and dynamically allocate data objects across multi-tier storage systems within multi-cloud environments. The framework integrates Long Short-Term Memory (LSTM)-based time-series forecasting for access prediction, supervised classification for hot/warm/cold tier assignment, and a reinforcement learning (RL) agent for adaptive storage placement decisions. Extensive experiments conducted using synthetic SAP workload traces — calibrated to documented SAP S/4HANA access pattern characteristics, including Zipf-distributed access frequencies and transactional-analytical-archival regime proportions — demonstrate up to 38% reduction in storage cost and 27% improvement in access latency compared to static tiering and rule-based baseline approaches. Scalability evaluations across dataset volumes ranging from 100,000 to one million data objects confirm that cost efficiency gains increase proportionally with data volume. The proposed framework demonstrates the viability of intelligent storage orchestration as a foundational capability for next-generation, cloud-native ERP deployments.
Downloads
References
D. Saxena, J. Kumar, A. K. Singh, and S. Schmid, "Performance analysis of machine learning-centered workload prediction models for cloud," IEEE Transactions on Parallel and Distributed Systems, vol. 34, no. 4, pp. 1313-1330, 2023. [Online]. Available: https://ieeexplore.ieee.org/document/10029931/
K. Nandkumar, R. Singh, and P. Sharma, "Accurate prediction of workloads and resources with multi-head attention and hybrid LSTM for cloud data centers," IEEE Transactions on Sustainable Computing, vol. 8, no. 3, pp. 412-425, 2023. [Online]. Available: https://ieeexplore.ieee.org/document/10077424/
H. Herodotou, H. Dong, and S. Babu, "Automating distributed tiered storage management in cluster computing," Proceedings of the VLDB Endowment, vol. 13, no. 1, pp. 43-56, 2019. [Online]. Available: https://doi.org/10.14778/3357377.3357381
L. Pang, A. Alazzawe, M. Ray, K. Kant, and J. Swift, "Adaptive intelligent tiering for modern storage systems," Performance Evaluation, vol. 160, p. 102332, 2023. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0166531623000020
T. Becker, R. Geihs, and F. Ramming, "Characterizing access patterns in enterprise ERP storage workloads," Journal of Systems and Software, vol. 196, p. 111546, 2023. [Online]. Available: https://doi.org/10.1016/j.jss.2022.111546
Z. Zheng, T. Zhu, Y. Liang, and K. Wang, "Adaptive data placement in multi-cloud storage: a non-stationary combinatorial bandit approach," IEEE Transactions on Cloud Computing, vol. 11, no. 4, pp. 3821-3834, 2023. [Online]. Available: https://ieeexplore.ieee.org/document/10223440/
H. Mao, M. Alizadeh, I. Menache, and S. Kandula, "Resource management with deep reinforcement learning," in Proc. 15th ACM Workshop on Hot Topics in Networks, 2016, pp. 50-56. [Online]. Available: https://dl.acm.org/doi/10.1145/3005745.3005750
Q. Liu, X. Liu, B. Dong, and F. Hao, "Deep reinforcement learning-based methods for resource scheduling in cloud computing: a review and future directions," Artificial Intelligence Review, vol. 57, no. 6, pp. 1-48, 2024. [Online]. Available: https://link.springer.com/article/10.1007/s10462-024-10756-9
A. Moschakis and H. Karatza, "A survey on deep reinforcement learning-based scheduling in distributed systems," Knowledge and Information Systems, vol. 66, no. 8, pp. 4741-4798, 2024. [Online]. Available: https://link.springer.com/article/10.1007/s10115-024-02167-7
N. Ghosh, S. K. Ghosh, and S. K. Das, "Distributed resource management in multi-cloud environments: challenges, state of the art and future research directions," Future Generation Computer Systems, vol. 129, pp. 315-330, 2022. [Online]. Available: https://doi.org/10.1016/j.future.2021.11.013
R. N. Calheiros, R. Ranjan, A. Beloglazov, C. A. F. De Rose, and R. Buyya, "CloudSim: a toolkit for modeling and simulation of cloud computing environments," Software: Practice and Experience, vol. 41, no. 1, pp. 23-50, 2011. [Online]. Available: https://doi.org/10.1002/spe.995
X. Zhang, Y. Wang, and L. Chen, "An improved LSTM-based prediction approach for resources and workload in large-scale data centers," IEEE Transactions on Cloud Computing, vol. 12, no. 2, pp. 501-514, 2024. [Online]. Available: https://ieeexplore.ieee.org/document/10486896/
T. Chen and C. Guestrin, "XGBoost: a scalable tree boosting system," in Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785-794. [Online]. Available: https://dl.acm.org/doi/10.1145/2939672.2939785
J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov, "Proximal policy optimization algorithms," arXiv:1707.06347, 2017. [Online]. Available: https://arxiv.org/abs/1707.06347
B. Schroeder, A. Merchant, and A. Vahdat, "Flash storage performance benchmarking for enterprise workloads," ACM Transactions on Storage, vol. 17, no. 2, pp. 1-28, 2021. [Online]. Available: https://doi.org/10.1145/3448706
M. Armbrust et al., "A view of cloud computing," Communications of the ACM, vol. 53, no. 4, pp. 50-58, 2010. [Online]. Available: https://doi.org/10.1145/1721654.1721672
G. Li, X. Wu, and Y. Zhang, "Multi-tier storage resource scheduling for cloud-based ERP: a cost-performance optimization model," Journal of Grid Computing, vol. 21, no. 3, pp. 1-19, 2023. [Online]. Available: https://doi.org/10.1007/s10723-023-09651-4
A. Rjoub, J. Bentahar, O. Abdel Wahab, and A. Bataineh, "Deep and reinforcement learning for automated task scheduling in large-scale cloud computing systems," Concurrency and Computation: Practice and Experience, vol. 33, no. 23, p. e5919, 2021. [Online]. Available: https://doi.org/10.1002/cpe.5919
S. Hochreiter and J. Schmidhuber, "Long short-term memory," Neural Computation, vol. 9, no. 8, pp. 1735-1780, 1997. [Online]. Available: https://doi.org/10.1162/neco.1997.9.8.1735
SAP SE, "SAP S/4HANA cloud: architecture and deployment guide," SAP Technical Documentation, 2023. [Online]. Available: https://help.sap.com/docs/SAP_S4HANA_CLOUD
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
