Self-Evolving Autonomous Software Architectures Using Large-Scale Graph Neural Networks and Real-Time Big Data Feedback Loops for Economic Optimization and Cost-Efficient Resource Allocation

Abstract

The rapid growth of cloud-native, microservice-based, and distributed computing environments has exposed the limits of conventional software architectures that rely on static rules, manually configured resource policies, and human-driven adaptation. These limitations often produce resource overprovisioning, higher operational costs, delayed failure recovery, and inefficient use of computational capacity. This study proposes a Self-Evolving Autonomous Software Architecture (SEASA) that combines large-scale Graph Neural Networks (GNNs) with real-time big-data feedback loops to enable continuous architectural learning, economic optimization, and cost-efficient resource allocation. The framework represents software ecosystems as dynamic heterogeneous graphs, where services, containers, databases, nodes, and infrastructure resources are modelled as interconnected entities, while dependencies, communication patterns, costs, and data flows are modelled as evolving edges. Runtime data were collected from Microsoft Azure Cloud workload traces, Google Cluster Workload Traces, and a synthetic microservice benchmark containing 500 services and 2.8 million interactions. Graph Convolutional Network, GraphSAGE, Graph Attention Network, and Temporal Graph Neural Network models were evaluated for anomaly detection, workload forecasting, architectural-state prediction, and autonomous adaptation. The TGNN-based framework achieved 97.4% accuracy, 96.8% precision, 96.1% recall, and a 96.4% F1-score, outperforming baseline models. It reduced adaptation latency by 38.7%, resource consumption by 24.5%, cloud operating costs by 21.8%, and overprovisioning by 27.3%. It also improved failure recovery, allocation efficiency, and service-level-objective compliance. These findings show that temporal graph intelligence and real-time economic feedback can transform software architectures into predictive, self-directed, and cost-aware computing ecosystems.

References

Ahmad, B., Jillani, S. A., Rafique, M., & Soomro, A. A. (2026, January). Design and monitoring of a tree-inspired vertical axis wind turbine for efficient urban wind utilization. In 2026 1st International Conference on Innovations in Information and Communication Technologies (IICT) (pp. 1–6). IEEE.

Ahmad, B., Majeed, M. K., Soomro, A. A., & Jillani, S. A. (2026, January). Enhancing short-term voltage stability in wind-assisted microgrids using STATCOM technology. In 2026 1st International Conference on Innovations in Information and Communication Technologies (IICT) (pp. 1–6). IEEE.

Alonso, J., Orue-Echevarria, L., Osaba, E., López Lobo, J., Martinez, I., Diaz de Arcaya, J., & Etxaniz, I. (2021). Optimization and prediction techniques for self-healing and self-learning applications in a trustworthy cloud continuum. Information, 12(8), 308. https://doi.org/10.3390/info12080308

Ali, S. I. (2026). Cognitive adversarial security neuro: Symbolic and generative AI defense ecosystems for self-evolving cyber resilience. In AI-driven cybersecurity systems, applications, and resilient infrastructure (pp. 139–162). IGI Global Scientific Publishing.

Alli, B. A., Yusuf, T. A., Ederhion, J., Otunlape, O., Koyenikan, O., Raheem, T. A., … Alli, Y. A. (2026). Self-evolving cyber defense: An analytical review of AI-driven autonomous and adversarial systems. Journal of Computer Virology and Hacking Techniques, 22(1), 59.

Anand, H., Khayambashi, K., Zandsalimi, Z., Taghizadeh, M., Hasnat, M. A., & Alemazkoor, N. (2025). Applications of graph neural networks in civil infrastructures: A review on transportation, power, water, and structural systems. Power, Water, and Structural Systems.

Baulin, V. A., Füchslin, R. M., Giacometti, A., Hauser, H., & Werner, M. (2025). Material-based intelligence: Self-organizing, autonomous and adaptive cognition embodied in physical substrates. arXiv. https://doi.org/10.48550/arXiv.2511.08838

Bisht, R., & Chaturvedi, N. (2026). Continual and self-healing cyber defence: An autonomous ML–RL architecture with lifelong learning, attack recovery, and model hardening. Attack Recovery, and Model Hardening.

Chennamsetty, C. S. (2024). Adaptive model training pipelines: Real-time feedback loops for self-evolving systems. International Journal of Advanced Research in Computer Science & Technology, 7(6), 11367–11373. https://doi.org/10.15662/IJARCST.2024.0706023

Cong, W., Tao, W., & Jinsong, B. (2026). Agentic digital twin-embedded maintenance methodology for energy equipment: A self-evolving operational paradigm. Journal of Manufacturing Systems, 84, 100–116.

Dhinakaran, K., Balavigneshwar, T. G., & Kavin, V. (2026, April). ZERO Mind: Self-evolving lifelong multimodal memory architectures for autonomous space digital twins. In 2026 IEEE 2nd International Conference on Quantum Photonics, Artificial Intelligence & Networking (QPAIN) (pp. 1–6). IEEE.

Fang, J., Peng, Y., Zhang, X., Wang, Y., Yi, X., Zhang, G., … Meng, Z. (2025). A comprehensive survey of self-evolving AI agents: A new paradigm bridging foundation models and lifelong agentic systems. arXiv. https://doi.org/10.48550/arXiv.2508.07407

He, J., Treude, C., & Lo, D. (2025). LLM-based multi-agent systems for software engineering: Literature review, vision, and the road ahead. ACM Transactions on Software Engineering and Methodology, 34(5), 1–30. https://doi.org/10.1145/3712003

Joshi, A., P, A., Prajapati, V., Anbalagan, S., & G, S. (2026). Federated spatio-temporal graph neural network for privacy-preserving vehicle trajectory prediction in autonomous driving. Frontiers in Artificial Intelligence, 9, 1856630. https://doi.org/10.3389/frai.2026.1856630

Kate, A. (2026). Secure software-defined networking (SDN) integration with zero trust for programmable critical infrastructure defense.

Li, Q., Ma, Y., Yu, J., Cao, S., Zhang, S., Zhang, P., & Yang, B. (2025). Intelligent fault diagnosis for HVDC systems based on knowledge graph and pre-trained models: A critical and comprehensive review. Energies, 18(24), 6438. https://doi.org/10.3390/en18246438

Lu, L., Liu, C., Zhang, C., Hu, Z., Lin, S., Liu, Z., … Chen, J. (2023). Architecture for self-evolution of 6G core network based on intelligent decision making. Electronics, 12(15), 3255. https://doi.org/10.3390/electronics12153255

Mogra, R., Thirugnanasambantham, K. G., Jain, S. K., & Bairagi, V. (2025, May). Bridging physics and intelligence: A comprehensive review on AI-driven multi-physics FEM for dynamic systems. In International Conference on Interactive Design and Digital Manufacturing (ICIDDM 2K25) (pp. 1–19).

Nasir, N., & Al Hamadi, H. (2025). Towards the neuromorphic Cyber-Twin: An architecture for cognitive defense in digital twin ecosystems. Frontiers in Big Data, 8, 1659757. https://doi.org/10.3389/fdata.2025.1659757

Paolini, D., Dini, P., Elhanashi, A., & Saponara, S. (2026). Advanced fault detection and diagnosis exploiting machine learning and artificial intelligence for engineering applications. Electronics, 15(2), 476. https://doi.org/10.3390/electronics15020476

Reddy, K. S. (2026). Next-generation cloud computing architectures with autonomous AI agents for enterprise service optimization. International Journal of Engineering & Extended Technologies Research (IJEETR), 8(2), 4776–4785.

Sapkota, R., Roumeliotis, K. I., & Karkee, M. (2025). UAVs meet agentic AI: A multidomain survey of autonomous aerial intelligence and agentic UAVs. arXiv. https://doi.org/10.48550/arXiv.2506.08045

Sathiri, D. (2026). Autonomous enterprise intelligence platforms enabled by collaborative multi-agent large language models. International Journal of Computer Science and Engineering Innovations, 2(3), 35–46.

Vijetha, B. (2026). Agentic intelligence for unified cyber defense: A self-adaptive framework for threat detection across cloud, edge, and IoT systems. IEEE Access, 14, 5104–5118.

Wang, J., Huang, N., Zhang, H., Liu, L., Fu, Q., Cao, K., … Jung, H. (2025). Self-learning model fusion for network anomaly detection: A hybrid CNN-LSTM-transformer framework. PLOS ONE, 20(10), e0332502. https://doi.org/10.1371/journal.pone.0332502

Wang, Q., Tian, W., Wang, R., Tsai, W. T., Shi, T., Liu, Z., & Xia, T. (2026, April). CogAgent: Self-evolving cognitive agents for multi-source fraud detection in heterogeneous financial networks. In Proceedings of the ACM Web Conference 2026 (pp. 3449–3458).

Xiang, Z., Yang, C., Chen, Z., Wei, Z., Tang, Y., Teng, Z., … Su, J. (2026). A systematic survey of self-evolving agents: From model-centric to environment-driven co-evolution.

Xu, Y., Zhang, W., Chen, Y., Lin, X., & Zhang, Y. (2026). Self-evolving agents as dynamic graph transformation: A survey and new perspective.

Yang, Y., Chai, H., Shao, S., Song, Y., Qi, S., Rui, R., & Zhang, W. (2025). AgentNet: Decentralized evolutionary coordination for LLM-based multi-agent systems. arXiv. https://doi.org/10.48550/arXiv.2504.00587

Zhang, Y., Zhao, X., Li, Z., Cheng, G., Yin, J., Zhang, L., & Chen, Z. (2024). Integrating artificial intelligence into operating systems: A survey on techniques, applications, and future directions. arXiv. https://doi.org/10.48550/arXiv.2407.14567