Managed Kubernetes
Automating, provisioning, scaling and managing containerised applications.
Simplify container management with Managed Kubernetes
Managed Kubernetes Service from Claranet
Operation of container-based applications on AWS, Azure, Google or the Claranet Cloud
Takeover of your complete lifecycle management
Many years of expertise with Kubernetes in complex container projects
What is Kubernetes?
Kubernetes is a management tool developed by Google as open source software for provisioning, managing and monitoring container environments. A container cluster manager like Kubernetes is necessary particularly in complex projects with a large number of containers spread across several servers or virtual machines. It takes care of processes like:
- assigning resources: e.g. Docker images to cluster nodes
- ensuring high availability: if a cluster node fails, the workloads are automatically started on another cluster
- load balancing: distribution of requests across all pods
What is a container?
Unlike virtual machines, containers do not emulate the hardware, but the operating system. Virtualisation thus takes place at a higher level, without a hypervisor. Several isolated containers can be operated within an operating system installation.
Containers contain both applications and the required operating system components (libraries). They provide the applications with a complete runtime environment. Kubernetes management software allows you to manage containers.
Kubernetes insights
Pods
Pods are the smallest deployable unit in Kubernetes. Think of them as a wrapper that holds one or more containers. A pod typically contains one container, in which one application or microservice runs. However, pods can also contain several containers if they need to access the same resources. Containers within a pod share the same host and therefore the same storage and network.
Nodes and clusters
Pods run on nodes, which are virtual or physical machines that are in turn grouped in clusters. A cluster possesses one or more control units called Kubernetes masters, which take care of managing the nodes. The master also controls communication with the kubelet agent, a process that is found on every node. For example, when a node fails, this initiates a restart of the affected pod on another node.
The most important features of Kubernetes include:
- Self-healing: When necessary, running containers are restarted based on defined rules.
- Auto-scaling: Automated adjustment of the required nodes based on load changes.
- Resource scheduler: Placement of containers/pods on the available nodes based on their resource requirements.
- Rolling upgrades: When applications are updated, the cluster automatically performs a rolling upgrade.
- Auto-discovery: If connections are detected between containers, services are automatically registered in the cluster-internal DNS zone.
- Secret/configuration management: Secure handling of passwords or API keys, for example.
- Load balancing: Incoming requests are distributed across all pods of a deployment based on selected strategies.
Managed Kubernetes Service for container environments
Claranet offers its customers a Managed Kubernetes Service for operating container-based applications. Claranet takes over the entire lifecycle management of a customer-dedicated cluster: design, integration, deployment, operation and maintenance. What is more, you benefit from numerous services like the provision of a dashboard, backup service or vulnerability check to increase security.
Let our experts advise you on sizing and the service features you want in an architecture workshop. We offer our Managed Kubernetes Service for the public clouds (Google, AWS, Azure) and for our own cloud.
