Portability Across High-Performance Edge Nodes

This screencast demonstrates how OpenNebula 7.2 enables seamless AI workload migration across heterogeneous HPC edge infrastructure. A vLLM inference service running the EuroLLM model is moved between two AMD EPYC-based nodes while preserving the same disk image and public WireGuard endpoint, showcasing flexible scaling, simplified infrastructure management, and service continuity for AI workloads at the edge. #IPCEICIS #8ra Funded by the #ONEnextgen UNICO IPCEI-CIS project: https://ONEnextgen.eu