Managed Kubernetes

Loopback operates your Kubernetes HA control plane and private network. Learn how scaling groups, upgrades, and central management support cluster operations.

5 min read
Managed Kubernetes

Kubernetes provides a shared foundation for running containerized applications. For a cluster to work reliably, its control plane, networks, nodes, and updates need to work together.

Loopback operates the highly available control plane and provides the private network. Your team chooses providers and locations, configures the required compute resources, and works with the unchanged Kubernetes API. Cluster management, access permissions, and billing come together on the same platform.

Operating the control plane

The control plane manages the cluster. It accepts configuration through the Kubernetes API and coordinates the implementation of the desired state.

For example, when you create a deployment, you describe which container image to use and how many replicas are needed. The control plane components then coordinate the required steps within the cluster.

For a team operating Kubernetes itself, this control layer belongs to its own operational responsibilities. With Loopback, the HA control plane is provided as a managed component of the cluster.

This reduces your team's responsibility for a central infrastructure task. Work on applications, resource requirements, and deployment configurations stays within the familiar Kubernetes model.

High availability and your application

A highly available control plane provides the foundation for managing the cluster. Application availability also depends on the application's own architecture.

This includes replica counts, resource planning, and restart behavior. Dependencies and the storage of application data also play a role.

Clear responsibilities help with planning:

ComponentOperational role
Control planeManages the cluster and processes changes through the API
NodesRun application workloads
Deployment configurationDescribes the application's desired state
NetworkEnables the intended connections
Persistent storageKeeps data independent of individual containers

Loopback provides the managed foundation. Your team designs how its applications run on top of it.

Integrate private networks

A Kubernetes cluster needs connections between its components. Infrastructure planning therefore includes internal communication alongside public access to applications.

Loopback provides the cluster's private network. It is managed alongside the cluster and forms part of the platform's operating model.

This lets you consider network configuration together with the other cluster resources. Your architecture determines which applications should be publicly accessible and which connections are needed internally.

Load balancers can distribute incoming connections in front of a project or directly at the Kubernetes Workspace.

Compute with scaling groups

Application requirements change. Additional services arrive, background processes need resources, or usage increases.

Loopback organizes a cluster's compute resources through scaling groups. You can adjust the auto-scale logic to workload requirements using several parameters.

A useful configuration requires your team to understand resource demand. This includes the baseline capacity needed and room for additional load.

Two layers work together:

  • At the application layer, you define how many instances of a workload should run.
  • At the infrastructure layer, enough capacity must be available to run those workloads.

Scaling groups address node capacity. Your application configuration determines which resources are needed on those nodes. Plan both layers together.

Upgrades during a maintenance window

Kubernetes updates belong to ongoing cluster operations. They need preparation and coordination with application requirements.

Loopback provides one-click upgrades and configurable maintenance windows. You can control updates through the platform and define an intended time frame.

A repeatable preparation process helps:

  1. Review changes: What changes and adjustments does the target version introduce?
  2. Review workloads: Are the resources and components in use suitable for the update?
  3. Try a test environment: Do the applications work with the intended version?
  4. Coordinate the maintenance window: When does the update fit your operational workflows?
  5. Check applications afterward: Are the expected functions and connections still working?

The platform simplifies carrying out the cluster upgrade. A coordinated process connects this technical step with application checks.

Choose providers and locations

Suitable infrastructure depends on the workload. Location requirements, compute demand, and data paths can differ between projects.

In Loopback, you choose providers and locations to suit the cluster. Loopback provides the HA control plane and private network. The underlying compute resources come from your selected providers.

Management stays on the same platform for your team. Even when infrastructure choices differ, you use the shared organization and permissions structure.

You can plan development and production clusters according to their respective requirements and operate them within the same management context.

The Kubernetes API remains in place

Loopback uses the standard Kubernetes API. You continue describing your applications through familiar Kubernetes resources.

Existing tools and workflows that use this API can remain in use. These include processes for provisioning deployments, services, and configurations.

An Admin Kubeconfig and an OIDC Kubeconfig are available to download for access.

The shared API gives your team a familiar working framework. Loopback takes on cluster operation tasks while application configuration stays within the Kubernetes model.

Manage clusters centrally

In Loopback, Kubernetes clusters fit into the organization, project, and Workspace structure. The Workspace represents the managed cluster.

This structure connects the technical resource with its organizational context:

LevelPurpose
OrganizationShared framework for members and permissions
ProjectGroups related resources
WorkspaceRepresents the Kubernetes cluster

Central Loopback RBAC lets you define who may use the respective product functions. Cluster management becomes part of the same access structure as Object Storage, Registry, and the other platform services.

Billing also comes together through Loopback. Infrastructure decisions across multiple providers remain within a shared management and billing model.

A foundation for applications

With Loopback, your team gets Kubernetes with a managed HA control plane, private network, and central cluster management. Scaling groups support capacity planning. One-click upgrades and maintenance windows structure updates.

You choose providers and locations to suit the workload. The standard Kubernetes API remains the interface for your applications and tools.

Your team can focus on application architecture and operations while Loopback operates the control plane.

Explore Kubernetes on Loopback

Your stack. One platform.

Start with managed Kubernetes, object storage, registry, or Bundles. One account, one permission model.