Today we’re excited to announce the launch of CKE, the easiest way to provision and manage Kubernetes clusters on your favorite cloud provider in just a few clicks. Traditionally, management of the Kubernetes cluster lifecycle has been difficult, especially when administering multiple highly-available clusters, potentially across a variety of regions and providers. CKE makes all of this simple from our intuitive UI.
What is CKE?
Containership Kubernetes Engine allows users to provision and manage Kubernetes clusters in a cloud-agnostic fashion, while adhering to the API standardization put in place by the Cloud Native Computing Foundation (CNCF). Since CKE is not a custom distribution, it does not preclude users from leveraging any existing native tools, such as kubectl. We have been working closely with the Kubernetes community to further the development of core Kubernetes building blocks related to cluster lifecycle management. CKE removes the complexity of dealing with those low-level building blocks, and provides users with a reliable and secure provisioning tool. Our mission was to build a tool which adheres to industry best-practices, while remaining flexible. As an interface, we’ve added a highly configurable cluster provisioning wizard to our Containership Cloud dashboard, as well as exposed the underlying RESTful provisioning APIs.
With a few clicks, users can easily spin up a best-practice cluster with sensible defaults. Alternatively, users can choose to configure more advanced options for their specific needs such as:
• Multiple node pool configuration
• Container networking solution
• Pod subnet CIDR range
• Metrics and logging
• Cloud controller manager integration (to automatically manage cloud-specific resources, such as load balancers and underlying storage)
• And many more in the near future
Why use CKE?
As previously mentioned, CKE is focused on bringing industry best-practices to provisioning Kubernetes clusters on any cloud provider. With Containership, it is easy to manage the lifecycle of any Kubernetes cluster, regardless of underlying provider, from a single pane of glass. By focusing on simplicity, we’ve built a tool that is operationally inexpensive, preventing cluster administrators from navigating through a fragmented choice of tooling to accommodate lifecycle management across clouds. For example, many provisioning tools are focused on integration with the “big three” cloud providers, while separate tooling is needed to create clusters on other providers that may better suit organizational needs. Even by leveraging a combination of tools, many operators are not sufficiently able to meet the agility required by their business. This results in either the concession of desired cloud topography, or increased operational burden on the infrastructure team.
To alleviate potential risks involved with navigating the confusing tooling ecosystem, many organizations opt to leverage managed Kubernetes services such as GKE or AKS. While these services decrease complexity, they are often too restrictive for common use cases. Conversely, master nodes on CKE are fully configurable as they are run on the user’s cloud provider. We believe this flexibility is imperative when choosing core infrastructure tooling. Additionally, through the Containership Cloud dashboard, users can manage all Kubernetes clusters in a consistent and standard manner - there is no need to learn and train engineers on disparate distributions from Google, Amazon and Azure.
With the increased velocity of CNCF project adoption, Containership remains dedicated to providing seamless integration with service meshes, monitoring solutions, log aggregators, and various other components. With CKE as the standard for cluster provisioning, we will provide a comprehensive platform that operators can use either as an off-the-shelf solution, or base platform for extension, that will be fully cloud agnostic and portable between underlying providers.
We’re extremely excited about the release of CKE, and wanted to get it into the hands of our users as soon as possible, but we still have a ton of additional features coming down the pipeline. Automatic node pool upgrades, cluster snapshotting and restore functionality across providers, and dedicated etcd provisioning are just a few examples. If there’s anything specific you’re interested in seeing, shoot us a message!
In the meantime, our initial release supports provisioning on DigitalOcean. Simply input your DigitalOcean API key and you are on your way. To learn more about creating a Kubernetes cluster on DigitalOcean click here. We will be adding support for a dozen cloud providers over the coming months. If you’re interested in learning how CKE works for private cloud installations of VMware and OpenStack, let us know.
Want to know more?
If you have any additional questions about CKE please reach out to us on Intercom or email at firstname.lastname@example.org