Date of Award
12-2025
Culminating Project Type
Thesis
Styleguide
ieee
Degree Name
Computer Science: M.S.
Department
Computer Science and Information Technology
College
School of Science and Engineering
First Advisor
Adriano Cavalcanti
Second Advisor
Jie Meichsner
Third Advisor
Bhaskar Ghosh
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Keywords and Subject Headings
Brain-Computer Interface (BCI), Deep Learning, GPU acceleration, Kubernetes, Machine Learning
Abstract
This thesis details the design, deployment, and evaluation of a Kubernetes-based infrastructure to support brain-computer interface (BCI) machine learning research in the Avatar Project at St. Cloud State University. EEG signals from OpenBCI headsets enable the control of drones and robots using deep learning; however, students lacked access to high-performance Graphic Processing Units (GPU) and centralized data management. To address this, a Kubernetes cluster was deployed across three on-premises servers, one GPU-enabled master and two CPU-only (Central Processing Unit) workers, integrated with JupyterHub for secure, browser-based access to notebook environments. Users selected CPU or GPU profiles for model training, with NFS-backed persistent storage ensuring data consistency and per-user workspace retention. Security and resource controls were enforced at the pod level. Results showed that GPU acceleration delivered a substantial 3.6x training speedup compared to CPU nodes, effectively reducing the necessary training time by over 70% on average. Persistent volumes safeguarded research progress, while the web platform broadened access to advanced resources. This solution bridges infrastructure gaps in academic BCI research, providing a scalable and equitable environment for collaborative machine learning.
Index Terms—Brain-Computer Interface (BCI), Deep Learning, GPU acceleration, Kubernetes, Machine Learning
Recommended Citation
Kuforiji, Simon I., "Kubernetes Scalable Deployment for Brain Computer Interface (BCI) Machine Learning" (2025). Culminating Projects in Computer Science and Information Technology. 62.
https://repository.stcloudstate.edu/csit_etds/62
Comments/Acknowledgements
I would like to thank the faculty and staff of the Department of Computer Science and Information Technology at St. Cloud State University for their immense support and guidance throughout my academic journey. I am very grateful to Dr. Adriano Cavalcanti for his mentorship throughout the project, and to Dr. Jie Meichsner, Dr. Ghosh Bhaskar and Dr. Bilal Al-Ahmad, the committee members, for their guidance.
I also want to express my deep appreciation to friends and family who have always been there for me every step of the way. I could not have achieved this without your support. Thank you.