Skip to documentation content
Cloud documentationCloud statusPilot in preparation
On this page
Get started

Plan your Cloud pilot

Turn a use case into a scoped proposal, a handover checklist and measurable acceptance criteria.

GuidePlanning & statusReading time · approx.8 minutesUpdated
Availability of this guidePilot planning

Use this planning guidance now for an inquiry and pilot proposal. Service commitments require a written agreement.

Before you start

Bring a workload description and a technical contact. You do not need your own cloud-provider account. Do not include credentials or production data in your inquiry.

1. Describe the workload

Start with one representative application. A useful pilot validates a real delivery and operating workflow, rather than just producing a running cluster. Tell us what the application needs and what would make the pilot successful for your team.

IncludeUseful detail
WorkloadContainer images, application ports, stateful components and dependencies such as a database or external API.
ScaleExpected replicas, CPU/memory requests, storage growth, traffic pattern and environments to separate.
RiskData sensitivity, acceptable interruption, recovery objectives and any restrictions on location or access.
OutcomeFor example: deploy your approved image, reach it over the agreed path and recover a test dataset.
Plan compute & storage →

2. Agree the scope before provisioning

ClusterPilot Cloud is in preparation. An inquiry requests an individual pilot proposal; it does not activate a subscription. Agree the following in writing before work starts. A pricing card alone does not confirm that a feature is qualified for your configuration.

  • ✓
    Cluster and node counts, compute characteristics, storage class, networking and the Kubernetes version to be provided.
  • ✓
    Platform fee, infrastructure items, taxes, billing interval and approval process for additional costs.
  • ✓
    Supported operations, maintenance window, support channel and which monitoring or backup services are included.
  • ✓
    Data handling, required agreements, pilot duration, acceptance criteria and export/deletion at the end.

Security & responsibilities → Understand pricing & usage →

3. Enter the Cloud workspace

  1. 01

    Receive a browser invitation

    Use the agreed sign-in and workspace invitation. ClusterPilot provides the managed access path; you do not receive provider credentials or a kubeconfig for the Cloud workflow.

  2. 02

    Check the workspace scope

    Confirm organization, workspace, plan, cluster limits and the named contacts shown in the proposal. A missing action means that the capability is not enabled for this pilot.

  3. 03

    Run the first browser workflow

    Start with a non-sensitive representative application, follow the operation state and record the acceptance result in the workspace. Do not switch to an untracked local command when the page reports a boundary.

4. Make the launch decision explicit

A healthy Pod is one check, not a production approval. Before a customer-facing launch, confirm application behavior under expected load, TLS and authentication, the escalation contact and the agreed data protection/recovery measures. Any unmet requirement needs an owner and a decision to resolve it or restrict the pilot.

Download the pilot checklist (.md) ↓

Use the Cloud console → Deploy an application with Helm in ClusterPilot →

Use this guide within your agreed pilot scope. Kubernetes examples are not evidence of a generally available Cloud capability.