Elastic Slurm: Scale NVIDIA AI and HPC Clusters on Demand | NVIDIA GTC Berlin 2026

In this screencast, we demonstrate how to deploy isolated clusters on demand, dynamically scale compute and GPU capacity, and release resources when workloads are complete.

By combining Slurm’s workload scheduling with OpenNebula’s cloud infrastructure management, OneSlurm enables more flexible, on-demand resource allocation for AI and HPC workloads.

Meet OpenNebula at NVIDIA GTC Berlin 2026: https://opennebula.io/events/nvidia-gtc-event/

#OpenNebula #OneSlurm #Slurm #NVIDIA #AIInfrastructure #HPC #GPUComputing