Karpenter is an open source Kubernetes autoscaler that automatically adds or removes nodes in real time. It reacts to instance changes within seconds and provisions the right compute CPU or GPU based workload on demand.
Run your autoscaling workloads on Exoscale’s GDPR compliant European cloud, trusted for reliability, privacy, and performance.
Run Kubernetes autoscaling on a fully European cloud, privacy focused, GDPR compliant, and hosted in data centers across Europe.
Karpenter is an add-on included in SKS PRO Cluster, you just need to enable it and it is fully supported by Exoscale SKS.
Run cloud-native applications of any size, from web services to AI training, with automatic CPU and GPU node scaling.
Use only the resources you need with scale-to-zero support and dynamic right-sizing. Save on infrastructure costs without sacrificing performance.
Simple billing, no hidden costs, and real experts available 24/7 across Europe.
Operate on a cloud powered by renewable energy, with modern data centers built for energy efficiency and long term sustainability.
Run large scale AI or ML pipelines efficiently with Karpenter as your Kubernetes autoscaler.
Automatically start GPU nodes for model training and inference, and scale them down when jobs are done.
You get high performance for your workloads and lower GPU costs thanks to on demand scaling.
Ensure consistent performance for customer facing services. Karpenter provisions the right compute resources in real time to handle traffic spikes or seasonal load, helping your Kubernetes scaling stay smooth and efficient without over provisioning.
For analytics, CI/CD, and background jobs, Karpenter optimizes node usage across your SKS cluster. Run compute heavy tasks when needed, then scale your nodes down automatically, reducing idle costs and improving overall cluster cost efficiency.
Ready to improve your Kubernetes scaling? Our team can help you enable Karpenter, optimize costs, and build efficient clusters on Exoscale’s European cloud. Contact us today to find the best setup for your workloads.
Contact usExpand your infrastructure with services that boost availability, optimize performance, and provide expert support for all your workloads. Exoscale offers everything you need to scale your projects successfully.
Built for compute heavy workloads like machine learning, data processing, and scientific simulations. Easily integrate GPU nodes into your Kubernetes clusters for accelerated performance where it matters most.
Discover
Deploy containerized applications on a production ready hosted and managed Kubernetes cluster in under two minutes. Use SKS as the orchestration layer for your application pods, using GPU virtual machine instances as underlying nodes.
Discover
Get the help you need to run your infrastructure with confidence through flexible support plans, designed to provide expert guidance, faster response times, and dedicated assistance tailored to your business.
DiscoverLearn how to get started with Kubernetes, enable Karpenter, and make the most of Exoscale’s managed Kubernetes platform.
Yes. Karpenter is fully integrated into Exoscale SKS PRO Cluster at no extra cost, ready to use without any manual installation.
Absolutely. Karpenter supports GPU autoscaling, ideal for AI, ML, and visualization workloads that need flexible GPU capacity.
Karpenter provisions or removes nodes in seconds, ensuring your Kubernetes clusters stay responsive and right-sized at all times.
Yes. Karpenter can automatically reduce node pools or clusters to zero when idle, helping you lower cloud costs.
No. Karpenter automatically selects the best Exoscale instance type for your workloads, CPU or GPU, based on resource requests.
You get advanced Kubernetes autoscaling with full data sovereignty, simple pricing, renewable powered data centers, and European based support. You get advanced Kubernetes autoscaling with full data sovereignty, simple pricing, renewable powered data centers, and European based support. Check the technical documentation.
