End-To-End Service Management in Multi-tenant Beyond 5G Networks

• Autores: Ángel M. Gama García
• Directores de la Tesis: Higinio Mora Mora (dir. tes.) , Jose María Alcaraz Calero (dir. tes.) , Qi Wang (dir. tes.)
• Lectura: En la Universitat d'Alacant / Universidad de Alicante ( España ) en 2024
• Idioma: español
• Número de páginas: 220
• Tribunal Calificador de la Tesis: Bo Wei (presid.) , Jose Pablo Salvador Garcia (secret.)
• Enlaces
  • Tesis en acceso abierto en: RUA
• Resumen

  • The thesis explores the management of end-to-end (E2E) services in Beyond 5G (B5G) multi-tenant networks. The increasing complexity of these networks, where multiple tenants share infrastructure, demands innovative solutions for resource allocation, security, and operational efficiency. Traditional methods are insufficient, particularly as B5G networks involve not only physical but also virtual layers of infrastructure. This research introduces new methodologies to dynamically optimize resource management, focusing on CPU and memory allocation, which are critical for maintaining network performance. One of the core contributions of this work is the development of tools for topology discovery and monitoring. The thesis proposes an architecture that enables accurate and continuous monitoring of network topologies within largescale, multi-tenant cloud infrastructures. These tools allow predictive analysis of software applications and system processes, which facilitates better operational management. The architecture includes the Process Inventory Agent (PIA), Process Monitoring Agent (PMA), Process Forecasting Agent (PFA), and Process Controlling Agent (PCA), each designed to enhance different aspects of service management. In addition to resource management, the thesis delves into cybersecurity, a crucial concern for B5G networks. The research develops proactive and reactive security mechanisms to mitigate cyber threats in industrial environments. This includes agent-based strategies designed to protect against both anticipated and real-time cyber attacks. The effectiveness of these tools has been validated through empirical evaluation, demonstrating their potential to maintain network performance in multi-tenant scenarios. Finally, the research highlights the scalability of the proposed solutions. Experimental results show that the tools developed in this project can handle increasingly complex network topologies with efficiency. This includes the ability to compute end-to-end network paths in networks with over 7,000 nodes in under 120 milliseconds. These results lay a strong foundation for future improvements in B5G network management, ensuring that B5G networks can meet the performance and security demands of modern communication systems. The research was supported by the 6G-BRAINS project, emphasizing its relevance to future developments in telecommunications.