Explainable Graph Neural Architecture Search for Adaptive Resource Management in Cloud-Native Service Meshes
Keywords:
explainable artificial intelligence, graph neural architecture search, cloud-native service meshes, adaptive resource management, service governance, infrastructure accountabilityAbstract
Cloud-native service meshes have become central infrastructures for managing microservice communication, observability, and policy enforcement at scale. However, resource management in such environments remains challenging because service interactions generate highly dynamic, heterogeneous, and non-Euclidean dependency structures. Graph neural networks offer a natural representation for these topologies, but manually designing graph models for adaptive resource control is slow and fragile under shifting workloads. Neural architecture search can automate model discovery, yet the resulting models often function as opaque decision systems, which creates governance and operational risks. This paper develops a system-level perspective on explainable graph neural architecture search for adaptive resource management in cloud-native service meshes. It examines structural trade-offs among search efficiency, graph representational capacity, inference cost, and interpretability. The discussion integrates service mesh control planes, sidecar data planes, telemetry pipelines, and policy engines to clarify how explainable graph architectures can be embedded into production infrastructures. The paper further analyzes deployment requirements, governance mechanisms, robustness, fairness, sustainability, and cross-domain lessons from traffic engineering, data center scheduling, and large-scale recommender systems. Rather than proposing a single algorithm, the paper argues that explainability must be treated as a first-class architectural constraint throughout the search, training, and operational lifecycle. This view shifts attention from predictive accuracy alone toward accountable, auditable, and resilient resource orchestration. The analysis concludes by outlining implications for future service mesh architectures, regulatory alignment, and sustainable cloud operations.
References
1. Barroso, L. A., Hölzle, U., & Ranganathan, P. (2018). The datacenter as a computer: Designing warehouse-scale machines. Synthesis Lectures on Computer Architecture, 13(3), 1-189.
2. Burns, B., Grant, B., Oppenheimer, D., Brewer, E., & Wilkes, J. (2016). Borg, Omega, and Kubernetes. Communications of the ACM, 59(5), 50-57.
3. Kipf, T. N., & Welling, M. (2017). Semi-supervised classification with graph convolutional networks. Proceedings of the 5th International Conference on Learning Representations.
4. Zoph, B., & Le, Q. V. (2017). Neural architecture search with reinforcement learning. Proceedings of the 5th International Conference on Learning Representations.
5. Elsken, T., Metzen, J. H., & Hutter, F. (2019). Neural architecture search: A survey. Journal of Machine Learning Research, 20(55), 1-21.
6. Li, Y., Yu, R., Shahabi, C., & Liu, Y. (2018). Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. Proceedings of the 6th International Conference on Learning Representations.
7. Mao, H., Alizadeh, M., Menache, I., & Kandula, S. (2016). Resource management with deep reinforcement learning. Proceedings of the 15th ACM Workshop on Hot Topics in Networks, 50-56.
8. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems 30, 4765-4774.
9. Doshi-Velez, F., & Kim, B. (2017). Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608.
10. Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021). A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 32(1), 4-24.
11. Veličković, P., Cucurull, G., Casanova, A., Romero, A., Liò, P., & Bengio, Y. (2018). Graph attention networks. Proceedings of the 6th International Conference on Learning Representations.
12. Liu, H., Simonyan, K., & Yang, Y. (2019). DARTS: Differentiable architecture search. Proceedings of the 7th International Conference on Learning Representations.
13. Xu, K., Hu, W., Leskovec, J., & Jegelka, S. (2019). How powerful are graph neural networks? Proceedings of the 7th International Conference on Learning Representations.
14. Wang, S., Tang, H., & Ouyang, J. (2023, July). A transformer-based neural architecture search method. In Proceedings of the Companion Conference on Genetic and Evolutionary Computation (pp. 691-694).
15. Gu, Z., Chang, R., Li, C., He, Q., Zhao, Z., & Tao, M. (2026, January). Uncertainty-aware hierarchical graph transformers for adaptive resource optimization in service meshes. In 2026 International Conference on Communication Networks and Machine Learning (CNML) (pp. 1351-1354). IEEE.
16. Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). Why should I trust you? Explaining the predictions of any classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 1135-1144.
17. Hamilton, W. L., Ying, R., & Leskovec, J. (2017). Inductive representation learning on large graphs. Advances in Neural Information Processing Systems 30, 1024-1034.
18. Ying, R., He, R., Chen, K., Eksombatchai, P., Hamilton, W. L., & Leskovec, J. (2018). Graph convolutional neural networks for web-scale recommender systems. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 974-983.
19. Lipton, Z. C. (2018). The mythos of model interpretability. Communications of the ACM, 61(10), 36-43.
20. Calcote, L., & Butcher, Z. (2019). The enterprise path to service mesh architectures. O'Reilly Media.
21. Beyer, B., Jones, C., Petoff, J., & Murphy, N. R. (2016). Site reliability engineering: How Google runs production systems. O'Reilly Media.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Advanced Artificial Intelligence Research
This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
