Kyriakos Chalvatzis, "Interactive user environment application on Kubernetes platform", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2025
https://doi.org/10.26233/heallink.tuc.103863
The following thesis presents the design and implementation of a user-oriented analytical computing system deployed on the Kubernetes platform. The system draws inspiration from core Unix design principles, particularly in file and directory permissions, user hierarchy and authorization management. The central ambition is to ease the operational quality of life for data analysts working with sizable or not datasets, while maintaining strict access control and support for concurrent multi-user interactions. Users can upload and share datasets, submit batch job into execution using pre-integrated applications, such as DuckDB, Bash, Octave, Pandas and others. Designed with modularity and extensibility in mind, the platform aims to abstract the complexity of orchestration, achieves abstraction of resource allocation and job scheduling, providing users with a seamless interface for analytical workflows. Each job lives and dies in a sandboxed containerized environment, ensuring reproducibility and isolation. The solution demonstrates the feasibility of delivering a lightweight, flexible platform that leverages Kubernetes’ native scalability to manage user resources, data, and jobs efficiently. It offers a pathway for democratizing access to powerful analytics tooling, especially in research and educational contexts where ease of deployment and extensibility are vital.