A unique way to launch pods and streamline
your K8s workload availability by leveraging multiple providers.
K8s Kosmos is a unique way to engage your multi-cloud strategy. We are the only provider to offer a managed Kubernetes that allows you to run pods across multiple cloud providers’ resources. You can host your pods on Scaleway Instances, or on resources from any other cloud provider of your choice, including private cloud servers.
Depending on your requirements and use cases, you can use any server, from any cloud provider, or even your Raspberry Pi, as Kubernetes nodes. Scaleway handles your Kubernetes management, and lets you choose the Instances that best meet your needs.
With Kubernetes Kosmos, we offer a superior level of redundancy for your applications, allowing them to become multi-AZ, multi-region and even multi-cloud provider. This also enables you to leverage cloud resources from multiple providers according to your needs.
Manage multiple providers within the same cluster and benefit from improved workload portability. Your teams will be able to move workloads across multiple cloud environments in just a few clicks.
Distribute your nodes between test and production environments across multiple cloud providers. Mix resources and Instances depending on your needs, run and manage clusters in single-access tools.
Avoid any drop in performance or incident coming from your CSP by creating cross-provider pools. Take advantage of the performance available with a multi-cloud architecture and ensure high availability for your services.
For a higher availability for your clusters, check out our Dedicated Control Plane
Deploy your applications across multiple cloud providers using their infrastructure resources. This unique solution allows you to mix Scaleway and external node pools in a single Kubernetes cluster, and manage all your pools with a single access point.
Scaleway allows you to attach nodes from any cloud provider to your managed K8s engine, making the management of your multi-cloud pools, nodes, and applications easier. You can spread your pods over different servers, depending on your application needs.
Duplicate your workload with a Scaleway pool to allow your applications to auto-scale. If a need arises on one of your applications, the Scaleway pool will take over to allow your workloads to scale, so you can spend your valuable time adapting the configuration of your cluster.
Mirror applications or split microservices across vendors. Bring high availability to a whole new level by replicating your deployments around the globe. You do not need to choose one provider, one region, or one data center to trust with your workload, you can distribute it everywhere and lower the risk of failure.
Benefit from Kuberentes Kapsule’s redundant control plane architecture, which has been successfully running thousands of production environments in General Availability since early 2019.
Discover also our Dedicated Control Plane for heavy workload and business critical applications
Your control plane and all of its components are redundant and replicated across our internal architecture cluster. Free yourself from the management of Kubernetes Cloud Controller Manager, K8s API server, and all Kubernetes services.
Join the Slack k8s Community and exchange with experts and enthusiasts!
Kubernetes Kosmos is the first managed Kubernetes engine allowing you to attach an Instance or a dedicated server from any cloud provider to Scaleway’s Kubernetes control plane. Based on Kubernetes Kapsule, pools containing Scaleway’s Instances will benefit from the same features, while external nodes from other providers will join the cluster to allow your containers to extend around the world.
From the Scaleway console, you can create Kubernetes pools that will run on either Scaleway Instances or pools on another provider’s Instances.
Then you can mix resources and Instances depending on your needs, run and manage clusters in single access tools. Please note that if you distribute nodes across different cloud providers, you may have to perform additional configurations to customize your node communication and ensure that the correct ports are open.
Multi-cloud clusters Kosmos control plane starts from 0,1444€/h ex.VAT
External nodes attached to a Kubernetes Kosmos cluster are completely free of charge. Whether you have just a few or hundreds of nodes within your multi-cloud cluster, your bill remains the same.
Auto-healing requires the Kubernetes control plane to be able to delete and create resources on your behalf. As this option is only available for Scaleway nodes, these operations are not possible with other providers. To ensure the health of nodes from other providers, you will need to implement auto-healing directly with the given provider.
Go to the “add a node” section of your Kubernetes cluster in the Scaleway Console where you will find the specific instructions to attach the node to your cluster. Remember that to do so, you will first need to create a multi-cloud pool using the console or the API.
On your other provider account, create the Instance or server you want to use in your cluster and follow the instructions on the Scaleway console to link your node to the cluster.
Be sure to follow the requirements listed on the Scaleway console while configuring external nodes. A misconfiguration will lead to the failure of the operation.
Yes. Any Instance or server running with the product’s prerequisites (as described in the Scaleway console) can be connected to your Kubernetes cluster. This includes dedicated servers, and also any external devices.
