Optimizing Cloud Networking Performance with Artificial Intelligence on Microsoft Azure
Keywords:
Azure Virtual Network Gateway, Artificial Intelligence, Cloud Networking Optimization, IKE Negotiation Analysis, hybrid cloud, azure, virtual network gateway (vng), artificial intelligence (ai), recurrent neural networks (rnns), traffic peaks, anomalies detection, resource allocation, azure network watcher, Wireshark, machine learning, latency reduction, throughput increase, a/b testing, cloud networking performanceAbstract
This paper offers an innovative strategy to improve the hybrid cloud performance over the Virtual Network Gateway (VNG) of the Microsoft Azure based on the latest Artificial Intelligence (AI). We use supervised Recurrent Neural Networks (RNNs) to predict traffic peaks and unsupervised Isolation Forests to detect anomalies in real time. Our AI-based framework will be optimized and used for resource allocation using the 4-week dataset of 100000 packets with timestamps, anonymized IPs, and headers, collected by Azure Network Watcher, Wireshark, and an Azure Log Analytics workspace on a particular date (15th October 2025). With Wireshark's deep filtering capabilities and Azure Machine Learning's powerful models, we can reduce latency, increase throughput, and filter out unwanted traffic. An extensive case study, backed by A/B testing and ablation tests, shows that VpnGw3AZ SKU throughput has been increased by 30-40 percent (95% CI: 28-42 percent) and that failure rates have been reduced by 25-35 percent (95% CI: 23-37 percent), and iPerf per Azure has confirmed these improvements.
References
[1] “Gartner Identifies the Top Trends Shaping the Future of Cloud.” Accessed: Jan. 18, 2026. [Online]. Available: https://www.gartner.com/en/newsroom/press-releases/2025-05-13-gartner-identifies-top-trends-shaping-the-future-of-cloud
[2] S. Senthil Pandi, P. Kumar, and T. A. Salman Latheef, “Cloud Platform Optimization Using Azure Machine Learning to Improve Performance and Reliability,” 2024 International Conference on Computational Intelligence for Green and Sustainable Technologies, ICCIGST 2024 - Proceedings, 2024, doi: 10.1109/ICCIGST60741.2024.10717550.
[3] “AI IN CLOUD COMPUTING: ENHANCING SERVICES AND PERFORMANCE.” Accessed: Jan. 18, 2026. [Online]. Available: https://www.researchgate.net/publication/385131309_AI_IN_CLOUD_COMPUTING_ENHANCING_SERVICES_AND_PERFORMANCE
[4] R. Ginting et al., “Hybrid cloud: bridging of private and public cloud computing,” iopscience.iop.orgG Aryotejo, DY KristiyantoJournal of Physics: Conference Series, 2018•iopscience.iop.org, p. 12091, 2018, doi: 10.1088/1742-6596/1025/1/012091/META.
[5] J. Hong, T. Dreibholz, J. A. Schenkel, and J. A. Hu, “An overview of multi-cloud computing,” SpringerJ Hong, T Dreibholz, JA Schenkel, JA HuWorkshops of the international conference on advanced information networking, 2019•Springer, vol. 927, pp. 1055–1068, 2019, doi: 10.1007/978-3-030-15035-8_103.
[6] N. T. of I. in E. research and undefined 2020, “Cloud computing security challenges,” researchgate.netNR TadapaneniInternational journal of Innovations in Engineering research and, 2020•researchgate.net, Accessed: Jan. 18, 2026. [Online]. Available: https://www.researchgate.net/profile/MustafaSabri3/publication/354788317_CLOUD_COMPUTING_SECURITY_CHALLENGES/links/614caea2a3df59440ba8a067/CLOUD-COMPUTING-SECURITY-CHALLENGES.pdf
[7] “About gateway SKUs | Microsoft Learn.” Accessed: Jan. 18, 2026. [Online]. Available: https://learn.microsoft.com/en-us/azure/vpn-gateway/about-gateway-skus
[8] D. Sirohi, N. Kumar, P. Rana, S. Tanwar, … R. I.-A. I., and undefined 2023, “Federated learning for 6G-enabled secure communication systems: a comprehensive survey,” SpringerD Sirohi, N Kumar, PS Rana, S Tanwar, R Iqbal, M HijjiiArtificial Intelligence Review, 2023•Springer, doi: 10.1007/S10462-023-10455-2.
[9] S. S. Gill et al., “AI for next generation computing: Emerging trends and future directions,” Internet of Things, vol. 19, p. 100514, Aug. 2022, doi: 10.1016/J.IOT.2022.100514.
[10] W. S. Al Farizi, I. Hidayah, and M. N. Rizal, “Isolation Forest Based Anomaly Detection: A Systematic Literature Review,” 2021 8th International Conference on Information Technology, Computer and Electrical Engineering, ICITACEE 2021, pp. 118–122, 2021, doi: 10.1109/ICITACEE53184.2021.9617498.
[11] S. Singh, D. Hoiem, D. F.-A. in neural, and undefined 2016, “Swapout: Learning an ensemble of deep architectures,” proceedings.neurips.ccS Singh, D Hoiem, D ForsythAdvances in neural information processing systems, 2016•proceedings.neurips.cc, Accessed: Jan. 18, 2026. [Online]. Available: https://proceedings.neurips.cc/paper_files/paper/2016/hash/c51ce410c124a10e0db5e4b97fc2af39-Abstract.html
[12] S. Troia, F. Sapienza, … L. V.-I. J. on S., and undefined 2020, “On deep reinforcement learning for traffic engineering in SD-WAN,” ieeexplore.ieee.orgS Troia, F Sapienza, L Varé, G MaierIEEE Journal on Selected Areas in Communications, 2020•ieeexplore.ieee.org, Accessed: Jan. 18, 2026. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/9279336/
[13] “AI Cost Optimization Strategies in Azure: Maximizing Efficiency.” Accessed: Jan. 18, 2026. [Online]. Available: https://snakconsultancy.com/blog/strategies-in-azure/
[14] N. Tanwar and Dr. P. K. K V, “Review on Machine Learning for Resource Usage Cost Optimization in Cloud Computing,” Int. J. Res. Appl. Sci. Eng. Technol., vol. 11, no. 5, pp. 468–472, May 2023, doi: 10.22214/IJRASET.2023.51489.
[15] S. Tuli, S. Gill, M. Xu, P. Garraghan, … R. B.-J. of S. and, and undefined 2022, “HUNTER: AI based holistic resource management for sustainable cloud computing,” ElsevierS Tuli, SS Gill, M Xu, P Garraghan, R Bahsoon, S Dustdar, R Sakellariou, O Rana, R BuyyaJournal of Systems and Software, 2022•Elsevier, Accessed: Jan. 18, 2026. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0164121221002211?casa_token=TDKMnKjnMQUAAAAA:2ipmTNp0NVdgFy6GKkvyYdfMqsFLUZgNk9qCwsAj4y9PGbRXVpRAX9yAMxipoVIYSorvXCUFjbT
[16] “VPN Gateway documentation | Microsoft Learn.” Accessed: Jan. 18, 2026. [Online]. Available: https://learn.microsoft.com/en-us/azure/vpn-gateway/
[17] S. Wu, N. Chen, A. Xiao, P. Z.-… S. & Tutorials, and undefined 2024, “Ai-empowered virtual network embedding: a comprehensive survey,” ieeexplore.ieee.orgS Wu, N Chen, A Xiao, P Zhang, C Jiang, W ZhangIEEE Communications Surveys & Tutorials, 2024•ieeexplore.ieee.org, Accessed: Jan. 18, 2026. [Online]. Available: https://ieeexplore.ieee.org/abstract/document/10587211/
[18] I. Ullah and Q. H. Mahmoud, “Design and Development of RNN Anomaly Detection Model for IoT Networks,” IEEE Access, vol. 10, pp. 62722–62750, 2022, doi: 10.1109/ACCESS.2022.3176317.
[19] A. Nanduri and L. Sherry, “Anomaly detection in aircraft data using Recurrent Neural Networks (RNN),” ICNS 2016: Securing an Integrated CNS System to Meet Future Challenges, Jun. 2016, doi: 10.1109/ICNSURV.2016.7486356.
[20] P. Vasiljevic, M. Matic, and M. Popovic, “Federated Isolation Forest for Efficient Anomaly Detection on Edge IoT Systems,” arxiv.orgP Vasiljevic, M Matic, M PopovicarXiv preprint arXiv:2506.05138, 2025•arxiv.org, pp. 30–35, Sep. 2025, doi: 10.1109/ZINC65316.2025.11103552.
[21] M. Yu, X. Z.-I. A. & S. Computing, and undefined 2023, “Anomaly Detection for Cloud Systems with Dynamic Spatiotemporal Learning.,” search.ebscohost.comM Yu, X ZhangIntelligent Automation & Soft Computing, 2023•search.ebscohost.com, vol. 37, no. 2, pp. 1787–1806, 2023, doi: 10.32604/IASC.2023.038798.
[22] F. T. Liu, K. M. Ting, and Z. H. Zhou, “Isolation forest,” Proceedings - IEEE International Conference on Data Mining, ICDM, pp. 413–422, 2008, doi: 10.1109/ICDM.2008.17.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Waseem John, Samyan Qayyum Wahla, Maheen Javed, Fatima Raees
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who submit papers with this journal agree to the following terms.
