Building a Government-Grade Orchestration Control Plane with Kestra
"Kestra gives us a modern orchestration platform we can run ourselves without compromising on governance."
The context
Dataport is one of Germany's leading public-sector IT services providers, designing, operating, and evolving shared digital platforms for federal states and government institutions across northern Germany.
As part of a broader cloud initiative, the team was building a private cloud service portal, a platform that would expose infrastructure services through APIs while meeting strict public-sector requirements.
Automation was essential. But in this environment, automation must be controlled, auditable, and self-hosted. External SaaS platforms were not an option, and ad-hoc scripting could not meet governance expectations.
They needed an orchestration layer that could act as a reliable control plane for long-running infrastructure workflows.
The tension: automation is required, but control cannot be delegated
The team faced a familiar public-sector dilemma.
On one side, demand for cloud services was increasing. On the other, execution had to remain compliant with internal standards, security requirements, and operational boundaries between teams.
Script-based automation and CI pipelines could trigger actions, but they didn’t provide a clear answer to fundamental questions:
- Who triggered this?
- What exactly ran?
- How is progress tracked?
- Where is the audit trail?
Automation without visibility quickly becomes operational risk.
Why Kestra
Kestra stood out because it aligned with how the organization already operates.
It can be fully self-hosted, deployed on OpenShift, and integrated with enterprise identity systems. Governance features like role-based access, separation between teams, and execution traceability are built in, not bolted on.
“Being able to build and manage workflows directly in the UI, while keeping everything on our own infrastructure, was a key differentiator.”
Platform Architect
Another decisive factor was usability. Workflows can be authored and evolved directly in the portal experience, reducing friction for platform teams while keeping execution controlled.
Kestra provided a balance that was difficult to achieve with scripts or traditional CI tooling: modern automation with enforceable governance.
The solution: a self-hosted orchestration control plane
Kestra was adopted as the orchestration layer for the private cloud platform.
The initial use case focuses on API-driven VM provisioning, but the architecture is designed to scale beyond a single workflow type. The team validated a pattern where:
- An external system creates and tracks a request
- Kestra orchestrates the long-running workflow
- Execution state and results are reported back for traceability
This keeps systems of record outside Kestra, while Kestra provides reliable execution, visibility, and control.
Event-driven integration plays a central role as well, with Kafka treated as a foundational requirement for triggering and connecting workflows.
What they run on the platform
Today, the platform is built around:
- API-driven orchestration workflows
- Kafka-based event handling
- Kubernetes-native execution on OpenShift
- PostgreSQL as the workflow backend
- S3-compatible internal object storage
- OIDC-based enterprise identity integration
This setup allows workflows to scale safely while maintaining clear operational boundaries between teams and services.
The outcome: controlled automation that can grow
The benefits are clear:
- The organization now has a single orchestration control plane aligned with public-sector constraints.
- Automation is observable and traceable by default, reducing operational ambiguity.
- Platform teams can deliver new workflows without bypassing governance.
- The foundation is in place to expand orchestration beyond VM provisioning without re-architecting later.
Or as the team summarized the win:
“Kestra lets us move forward with cloud automation, while staying fully in control.”
Platform Team
What’s next
After validating the core orchestration patterns, the team plans to extend Kestra to additional workflows within the private cloud platform.
The long-term goal is clear: a standardized orchestration control plane that government teams can rely on for secure, auditable cloud operations.
What would change if your public-sector cloud automation ran through a self-hosted orchestration control plane, built for governance, visibility, and scale?
Experience Kestra Today
What would change if your public-sector cloud automation ran through a self-hosted orchestration control plane, built for governance, visibility, and scale?
