Deploying AI Inferencing Workloads on an Ampere ARM64-based Edge Cluster

This screencast highlights deploying a Ray appliance from the OpenNebula Marketplace, designed for ARM64 and pre-configured with Hugging Face’s QWEN 2.5 model.

The appliance is instantiated on an Ampere ARM host, and the demo concludes with a live interaction using the built-in ChatBot. The setup runs on a 2-node Ampere ARM64 edge cluster deployed with OpenNebula’s OneDeploy tool powered by Ansible.

Perfect for exploring AI inference, edge computing, and ARM64 infrastructure-as-code.

#edgecomputing #ARM64 #Ampere #OpenNebula #infrastructureascode #OneDeploy #Ansible #AIinference #HuggingFace #RayFramework #QWEN #edgeAI #IPCEICIS #8ra #ONEnextgen

Funded by the ONEnextgen UNICO IPCEI-CIS project:
http://ONEnextgen.eu

Видео Deploying AI Inferencing Workloads on an Ampere ARM64-based Edge Cluster канала OpenNebula