Autoscaling GPU Clusters Anywhere — Hyperscalers, Neoclouds & Ba...

on YouTube ↗06m

" summary

vCluster runs a full Kubernetes control plane in a single pod and integrates Karpenter to autoscale GPUs across hyperscalers, neoclouds and bare metal, claiming to cut idle GPU time and enable cloud bursting. Implementation stitches v-cluster auto nodes with Terraform/OpenTofu, ClusterAPI, KubeVirt and NVIDIA BCM, supports provisioning from multiple providers, lifecycle (upgrade/decommission) and cost-driven consolidation, and ships a reference architecture for NVIDIA DGX superpods that the demo reports working end-to-end.

AI-written summary · about

" related talks

• 08mContainerization – It’s Not Just for CPUs Anymore: Introducin...Nathan Goulding · Platform Engineering Day NA 2024
• 12mCluster API To Streamline Kubernetes Clusters Lifecycle Management...Megha Singhal, Gulshan Khatri · KubeCon + CloudNativeCon India 2025
• 12mKubespray: Driving Cost-Efficiency for AI on KubernetesAntoine Legrand, Mohamed Zaian · KubeCon + CloudNativeCon Europe 2025
• 27mScaling K8s Clusters Without Losing Your Mind or MoneyYael Grossman · Conf42 Cloud Native 2023
• 27mHarness the power of Karpenter to scale, optimize & upgrade K8sChakkree Tipsupa · Conf42 Kube Native 2024
• 31mAccelerate AI adoption: automated cloud efficiencyRoss Gustavson · SREday San Francisco 2024 Q4
• 21mWhy should you bother about cluster autoscaling- KarpenterRaja Ganesan · Conf42 Cloud Native 2022
• 01h 15mTutorial: Build, Operate, and Use a Multi-Tenant AI Cluster Based Entirely on Open SourceC. Misale, O. Tardieu, D. Grove · KubeCon + CloudNativeCon Europe 2025