This work investigates the computational footprint of containerized 5G components deployed on a virtualized infrastructure, and leverages this profiling to propose an intelligent orchestration mechanism to maximize both the capacity offered to mobile network users and resource availability to application level services deployed on the computing infrastructure. We deployed and configured containerized 5G core network and radio access components using Open5GCore and UERANSIM, analyzing their resource occupation on the host server. The results show that the User Plane Function and the virtualized gNodeB introduce the most significant processing overhead. For instance, as the system scales from 5 users with 5 Mbps channels to 100 users with 50 Mbps channels, we observe an increase in CPU utilization of approximately 522%, highlighting the strong dependency between traffic load and resource consumption. Then, we incorporate the measured resource profiles into a state-of-the-art service orchestration system capable of deploying both 5G core network components and application-level services on heterogeneous computing equipment. Using an appropriate simulation tool, we evaluate our intelligent scheduling algorithm for service placement and compare it with relevant heuristic strategies. Results demonstrate that the proposed approach reduces service blocking probability by up to 76.77% while improving the bandwidth delivered to users by up to 46.76%, depending on service arrival rates and the number of users connected to the network. Based on these findings, we conclude that the proposed approach proves helpful in improving both resource efficiency and service quality.
Gaetano Francesco Pittalà, G. Davoli, Walter Cerroni et al.· IEEE Conference on Network S...· 0 citations
The evolution towards 6G edge-cloud ecosystems demands autonomous, intent-based network management to handle unprecedented infrastructure complexity. While Large Language Models offer promising capabilities for translating high-level user intents into network configurations, current monolithic approaches suffer from cognitive overload, hallucinations, and a profound inability to safely execute low-level data plane mutations. To bridge this gap, we introduce Edgent, a novel framework that integrates hierarchical Agentic AI with Extended Berkeley Packet Filter technologies via the Model Context Protocol. Edgent utilizes a state-driven Supervisor, enhanced by Retrieval-Augmented Generation, to decompose abstract human intents into deterministic execution graphs and dynamically delegate tasks to domain-specific worker agents. We empirically validate the framework by autonomously deploying a distributed, in-kernel DDoS mitigation pipeline across scaled containerized topologies containing up to 85 nodes. Extensive evaluations demonstrate high orchestration reliability; notably, even heavily quantized Small Language Models (e.g., 4B parameters) achieve near-perfect zero-shot execution and 100% overall task completion through autonomous error recovery. Finally, latency and resource profiling confirm that the multi-agent framework can be efficiently driven by fully localized models compatible with orchestration tasks directly within resource-constrained edge environments, therefore this work positions Edgent as a pragmatic step toward the realization of zero-touch nextgeneration networks.
Raffaele Di Tommaso, G. Davoli, Pietro Spadaccino et al.· IEEE Conference on Network S...· 0 citations