Karpenter
Kubernetes Wasn’t Built for GPUs. Make It Behave
Kubernetes counts whole GPUs and treats pods as disposable. An LLM pod is neither. Share the silicon with MIG/MPS/time-slicing and stop paying for idle ...
The Ultimate Guide to GPU Scaling With Karpenter
Karpenter GPU scaling on Amazon EKS: avoid common mistakes, optimize Spot capacity, reduce cold starts and improve utilization for AI workloads ...
What to Expect From Kubernetes 1.36
Kubernetes 1.36 launches April 22, 2026, marking a major shift in networking as Ingress-Nginx retires in favor of the more scalable Gateway API. Key updates include bolstered Linux User Namespaces for better ...
vCluster Embraces Karpenter for Dynamic Scaling of Virtual Kubernetes Nodes
vCluster Labs today added an ability to automatically scale nodes running on a virtual Kubernetes cluster. Company CEO Lukas Gentele said the Auto Nodes capability added to the vCluster platform are enabled ...
Karpenter Carves Out Keener Kubernetes Compute Consumption
Open source Kubernetes cluster auto-scaling and compute management project Karpenter dons its graduation robes this month as it moves to its 1.0.0 version release ...
