Adaptive Scaling Architecture with Learning Support

• André Luiz de J. Gonçalves UFG
• Leandro A. Freitas IFG https://orcid.org/0000-0001-9801-2867
• Antonio Oliveira-Jr UFG / Fraunhofer Portugal AICOS https://orcid.org/0000-0003-3595-3525

DOI: https://doi.org/10.5753/sbrc.2025.5933

Abstract

Meeting performance and stability demands in latency-sensitive applications is one of today’s major technological challenges. This work presents an Adaptive Scaling Architecture with Learning Support, applied to the context of immersive applications and based on the combination of hardware metrics and application-level events to optimize resource allocation. The implementation uses Kubernetes and the Kubernetes Event-driven Autoscaler (KEDA), with the Hubs VR application as a case study. Experiments were conducted resulting in the construction of two structured datasets: one based solely on hardware metrics and another also integrating application events. These datasets represent a relevant outcome of the research, serving as a foundation for analyses and the development of predictive strategies. The results indicate that combining metrics can lead to more agile and stable responses to load variations, contributing to the advancement of adaptive solutions in dynamic environments.

Keywords: Adaptive scaling, Automatic scaling, Immersive applications, MAPE-K

References

Bartolomeo, G., Cao, J., Su, X., and Mohan, N. (2023). Characterizing distributed mobile augmented reality applications at the edge. In Companion of the 19th International Conference on emerging Networking EXperiments and Technologies, pages 9–18.

Benmerar, T. Z., Theodoropoulos, T., Fevereiro, D., Rosa, L., Rodrigues, J., Taleb, T., Barone, P., Tserpes, K., and Cordeiro, L. (2023). Intelligent multi-domain edge orchestration for highly distributed immersive services: an immersive virtual touring use case. In 2023 IEEE International Conference on Edge Computing and Communications (EDGE), pages 381–392. IEEE.

Cheng, K., Zhang, S., Tu, C., Shi, X., Yin, Z., Lu, S., Liang, Y., and Gu, Q. (2023). Proscale: Proactive autoscaling for microservice with time-varying workload at the edge. IEEE Transactions on Parallel and Distributed Systems, 34(4):1294–1312.

Dimolitsas, I., Spatharakis, D., Dechouniotis, D., Zafeiropoulos, A., and Papavassiliou, S. (2023). Multi-application hierarchical autoscaling for kubernetes edge clusters. In 2023 IEEE International Conference on Smart Computing (SMARTCOMP), pages 291–296. IEEE.

Dogani, J., Namvar, R., and Khunjush, F. (2023). Auto-scaling techniques in container-based cloud and edge/fog computing: Taxonomy and survey. Computer Communications, 209:120–150.

Gonçalves, A. L. d. J., Oliveira-Jr, A., and Freitas, L. A. (2023). Revisao sistemática das aplicaçoes imersivas com base nas tecnologias habilitadoras b5g/6g, mec e ia. Anais da XI Escola Regional de Informática de Goiás.

Grafana, 2024. Grafana: The open and composable observability platform. [link]. Acesso em 21 de janeiro de 2024.

Han, B., Pathak, P., Chen, S., and Yu, L.-F. C. (2022). Comic: A collaborative mobile immersive computing infrastructure for conducting multi-user xr research. IEEE Network.

Hubs Foundation (2024). Hubs foundation - open source social vr platform. Accessed: 2024-07-12.

k6.io (2024). k6 Open Source: A modern load testing tool for developers. Acessado em 20 de dezembro de 2024.

KEDA Project (2024). Keda: Kubernetes-based event driven autoscaler. Acessado em: 10-07-2024.

Kubernetes (2022). Kubernetes documentation.

LABORA-INF-UFG (2025). Repositório do projeto adaptscale. [link].

López-Ramírez, G. A., Aragón-Zavala, A., and Vargas-Rosales, C. (2024). Exploratory data analysis for path loss measurements: Unveiling patterns and insights before machine learning. IEEE Access.

Malburg, L., Hoffmann, M., and Bergmann, R. (2023). Applying mape-k control loops for adaptive workflow management in smart factories. Journal of Intelligent Information Systems, 61(1):83–111.

Mehta, R., Sahni, J., and Khanna, K. (2023). Task scheduling for improved response time of latency sensitive applications in fog integrated cloud environment. Multimedia Tools and Applications, 82(21):32305–32328.

Nguyen, T.-T., Yeom, Y.-J., Kim, T., Park, D.-H., and Kim, S. (2020). Horizontal pod autoscaling in kubernetes for elastic container orchestration. Sensors, 20(16):4621.

Pandas (2025). Pandas: Python data analysis library. [link]. Accessed: 2025-01-13.

Pelle, I., Czentye, J., Dóka, J., and Sonkoly, B. (2019). Towards latency sensitive cloud native applications: A performance study on aws. In 2019 IEEE 12th International Conference on Cloud Computing (CLOUD), pages 272–280. IEEE.

Prometheus, T. (Acesso em 21 de janeiro de 2024). Prometheus: The Comprehensive Monitoring and Alerting Toolkit. [link].

Quattrocchi, G., Incerto, E., Pinciroli, R., Trubiani, C., and Baresi, L. (2024). Autoscaling solutions for cloud applications under dynamic workloads. IEEE Transactions on Services Computing.

Santos, J., Wauters, T., Volckaert, B., and De Turck, F. (2023). gym-hpa: Efficient auto-scaling via reinforcement learning for complex microservice-based applications in kubernetes. In NOMS 2023-2023 IEEE/IFIP Network Operations and Management Symposium, pages 1–9. IEEE.

Siriwardhana, Y., Porambage, P., Liyanage, M., and Ylianttila, M. (2021). A survey on mobile augmented reality with 5g mobile edge computing: Architectures, applications, and technical aspects. IEEE Communications Surveys & Tutorials, 23(2):1160–1192.

Taleb, T., Boudi, A., Rosa, L., Cordeiro, L., Theodoropoulos, T., Tserpes, K., Dazzi, P., Protopsaltis, A. I., and Li, R. (2022). Toward supporting xr services: Architecture and enablers. IEEE Internet of Things Journal, 10(4):3567–3586.

Wu, D., Yang, Z., Zhang, P., Wang, R., Yang, B., and Ma, X. (2023). Virtual-reality inter-promotion technology for metaverse: A survey. IEEE Internet of Things Journal.

Yuan, H. and Liao, S. (2024). A time series-based approach to elastic kubernetes scaling. Electronics, 13(2):285.