The 6G‑Cloud project participated in the 21st International Conference on Network and Service Management (CNSM 2025), held in Bologna, Italy, from 27–31 October 2025. The conference, themed ‘AI and Sustainability in the Future of Network and Service Management,’ brought together leading researchers and industry experts to discuss innovations in communication-system, network, and service management for next-generation technologies.

Solomon Wassie from the University of Bern presented the paper ‘MARC-6G: Multi-Agent Reinforcement Learning for Distributed Context-Aware SFC Deployment and Migration in 6G Networks.’ The research addresses the challenges of orchestrating Service Function Chains (SFCs) in the Cloud Continuum Framework (CCF), which extends computing capabilities across near-edge, far-edge, and extreme-edge nodes to meet the performance demands of emerging 6G applications.

The paper introduces MARC‑6G, a decentralised, multi-agent reinforcement learning framework that enables agents to monitor different network segments, share experience, and make local orchestration decisions. By collaboratively optimising control policies, MARC‑6G enhances global SFC deployment efficiency, reduces migration costs, and lowers energy consumption. Simulation results indicate 34% lower migration costs and 12.5% lower energy consumption compared to centralized DRL baselines, demonstrating the scalability and efficiency of this distributed approach.

CNSM 2025 provided a valuable platform for 6G‑Cloud to showcase its work on AI-driven, cloud-native orchestration and engage with the international research community on scalable and sustainable 6G network management solutions.