Hands-On Network Slicing Optimization: A Graph Neural Network Perspective

Yassine Hadjadj-Aoul

IRISA, University of Rennes 1, France

20th November 2024, Wednesday (10:30-12:10), Room 1

Session Chair : Zibouda Aliouat, Setif 1 University, Algeria

ABSTRACT

Network slicing has emerged as one of the most crucial technologies for achieving full network automation. By dividing a physical network into multiple virtual slices, each tailored to meet specific service requirements, network slicing allows operators to manage complex, heterogeneous services over shared infrastructure. However, the management and optimization of these slices present a significant challenge due to the dynamic nature of the network and the need for real-time adaptation to diverse service demands. This complexity is often challenging for traditional optimization methods that do not scale or adapt efficiently as network conditions change. Recent advances in machine learning, especially reinforcement learning, hold much promise for automatic resource management. Nevertheless, these approaches ignore the rich underlying graph structure associated with network slicing, which negatively affects their efficiency and generalization capabilities. This tutorial introduces Graph Neural Networks (GNNs) as a powerful tool to overcome these limitations and handle in an effective way graph-structured data inherent to network slicing. By leveraging GNNs, we optimize the placement of virtual services and resource allocation, making the automation of network slices both scalable and adaptive. It will also provide practical insights into how GNNs can be applied to solve real-world network slicing challenges, alongside a discussion of opportunities and limitations of these methods. Further, this tutorial will provide future research directions that may help unlock the full potential of GNNs in driving network automation for 5G and beyond networks.

BIOGRAPHY

Prof. Yassine Hadjadj Aoul is currently working as a full professor at the University of Rennes 1, France, where he is also a member of the IRISA Laboratory and the INRIA project-team Dionysos. He received a B.Sc. In computer engineering with high honours from Mohamed Boudiaf University, Oran, Algeria, in 1999. Dr. Hadjadj received his Master's and Ph.D. degrees in computer science from the University of Versailles, France, in 2002 and 2007, respectively. He obtained the qualification to direct research (HDR) in 2021. He was an associate professor at the University of Rennes from 2009 to 2022. He was an assistant professor at the University of Versailles from 2005 to 2007, where he was involved in several national and European projects such as NMS, IST-ATHENA, and IST-IMOSAN. He was also a post-doctoral fellow at the University of Lille 1 and a research fellow, under the EUFP6 EIF Marie Curie Action, at the National University of Dublin (UCD), where he was involved in the DOM'COM and IST-CARMEN projects, which aim at developing mixed Wi-Fi/WiMAX wireless mesh networks to support carrier grade services. His main research interests concern the fields of wireless networking, multimedia streaming architectures and protocols, congestion control protocols and QoS provisioning, and satellite and space communications. Dr. Hadjadj has been on the technical program committee of different IEEE conferences, including Globecom, ICC, VTC, PIMRC and IWCMC. His work on multimedia and wireless communications has led to more than 50 technical papers in journals and international conference proceedings.