Trinium documentation
The help Trinium shows inside the builder, published. Start with the install guide, or look up a node.
One Linux server, Docker and about ten minutes. The commands are on one page.
How Trinium worksPipelines, test runs, versions and deploys, Flows, schedules, parameters, connections, roles and the audit trail.
Install and administrationSystem requirements, configuration, HTTPS, upgrades, backups, more servers, disk, logs and troubleshooting.
NodesEvery node on the canvas: what it does, its inputs and outputs, when to use it, examples and tips.
ConnectorsEvery system Trinium connects to: what each connection asks for, what its Test checks, and the nodes that use it.
Expression languageOne SQL expression language everywhere you type a rule: the rules, the functions and worked recipes.
Moving to TriniumGuides for moving work to Trinium from the tools you use today, tool by tool.
Release notesWhat changed in each release, newest first.
What Trinium is
Trinium is self-hosted visual ETL and data integration. You build a pipeline on a canvas in your browser: nodes read data, shape it and write it out, and each node streams its rows to the next as they arrive. A Flow runs pipelines in order and decides what happens when one fails. Every pipeline has a draft you edit, numbered versions, and separate Staging and Production deployments. All of it runs on your own Linux servers.
How these pages are made
Every node page is generated from the help text the builder itself shows, so the page and the product say the same thing. Each page names the Trinium release it describes.