Multi-cloud Kubernetes Redefined¶
Build multi-cloud, hybrid-cloud, and on-premises Kubernetes clusters that mix and match nodepools across AWS, GCP, Azure, Hetzner, across your own bare-metal servers, and more, managed through a single InputManifest.
What is Claudie¶
Claudie is a platform for managing multi-cloud, hybrid-cloud, and on-premises Kubernetes clusters. In a single cluster you can mix and match nodepools from various cloud providers and your own on-premises or bare-metal servers, e.g. a nodepool in AWS, another one in GCP, and a third one running on hardware in your own datacenter. This is our opinionated way to build multi-cloud, hybrid-cloud, and on-premises Kubernetes infrastructure. On top of that, Claudie supports Cluster Autoscaler on the managed clusters. See the On-Premises provider docs for how to attach your own machines.
Supported Providers¶
Use cases¶
Claudie has been built as an answer to the following Kubernetes challenges:
- Cost savings
- Data locality & compliance (e.g. GDPR)
- Managed Kubernetes for providers that do not offer it
- Cloud bursting
- Service interconnect
- On-premises & hybrid integration - extend existing bare-metal, co-located, or private-datacenter servers into a Kubernetes cluster alongside cloud nodepools
- GPU & AI/ML workloads - Provision GPU nodes across multiple clouds, run NVIDIA GPU Operator, and leverage Cluster Autoscaler for GPU-aware scaling
You can read more here.
Features¶
- Manage multi-cloud, hybrid-cloud, and on-premises Kubernetes clusters
- Management via IaC
- Fast scale-up/scale-down of your infrastructure
- Loadbalancing
- Persistent storage volumes
- On-premises & bare-metal nodepools - integrate your existing servers as first-class cluster nodes via SSH (details)
- GPU workload support - Deploy GPU-accelerated nodepools with NVIDIA GPU Operator compatibility and autoscaler-aware GPU scheduling
See more in How Claudie works sections.
AI/ML Ready Infrastructure
Claudie lets you build GPU clusters that span multiple cloud providers, combining cost-effective GPU instances from different vendors into a single Kubernetes cluster. It comes with the NVIDIA GPU Operator and Cluster Autoscaler for scalable production-ready AI/ML workloads.
Bring Your Own Hardware
Already running on-premises or co-located servers? Claudie treats them as first-class nodepools. Add them to any cluster alongside AWS, GCP, Azure, and other cloud nodepools using the On-Premises provider, all from the same InputManifest.
What to do next¶
In case you are not sure where to go next, you can just simply start with our Getting Started Guide or read our documentation sitemap.
If you need help or want to chat with us, feel free to join our slack channel
(Get invite here)