What is stardelt
stardelt is an open-source data platform that runs on your own Kubernetes cluster. It installs proven open-source projects and connects them, so you get one sign-in, one interface and one API for storing, querying, scheduling and showing data. See a team's day in five steps.
Nova, stardelt's interface: catalogs, recent loads and queries, and the services, in one place.
Pre-alpha
APIs, resources and install steps change without notice. There are no tagged releases yet. Try it, but do not run your business on it.
What you can do with it
- Query Iceberg tables with SQL in the browser, from a notebook or from a script.
- Load files, APIs and databases into tables, and keep source credentials out of your code.
- Schedule SQL and notebooks, and publish pipelines from Git.
- Show results in dashboards, and control who sees what by group.
- See what ran, how long it took, why it failed and which tables it touched.
- Automate all of it with an API, a Python package, a command line and AI assistants.
Why you might want it
- Your data stays with you. Everything runs in your cluster. stardelt's own components make no required outbound calls, and there is no license server. See Sovereignty.
- You can leave. Tables are open Iceberg and Parquet files on storage you choose.
- No license surprises. Permissive licenses throughout, with one documented exception. See the license policy.
- Standard parts. Trino, Iceberg, Airflow, Superset and JupyterHub are what they are upstream, so what you already know still applies.
stardelt may not suit you if you want a managed service or do not want to operate Kubernetes.
Where to go next
| You want to | Read |
|---|---|
| See it working | Quickstart |
| Understand the parts | How it works |
| Install it for real | Prerequisites |
| Work with data | From source to dashboard |
| Run it | Running stardelt in production |
| Help build it | Developer guide |
| Know what comes next | Roadmap |