Guides¶

These guides walk you through Vulcan's core workflows. Each one solves a specific problem, with real examples from the Orders360 project.

If you're new, start at the top. If you're looking for something specific, jump to the section you need.

Set up and first run¶

Get Started - Install Vulcan, spin up a local Postgres environment with Docker, and run your first project. By the end, you'll have a working stack on your machine.

Data Product Lifecycle - The full path from make up to production. Covers every phase: infrastructure setup, model creation, testing, semantic layer configuration, and deployment.

Day-to-day workflows¶

Plan - vulcan plan shows you what changed, what gets recomputed, and what gets promoted before anything takes effect. Use this before every deployment.

Run and Scheduling - plan applies changes. run processes new data intervals. This guide covers the difference, shows you how vulcan run works, and walks through cron-based scheduling for production.

Models - Add, edit, evaluate, and manage SQL and Python models. Covers the full workflow: creating a model, previewing its output with vulcan evaluate, and applying it with a plan.

Going deeper¶

Incremental by Time - Full refreshes reprocess everything. If you have a year of sales data, that's 365 days recomputed on every run. Incremental models process only the new intervals. This guide shows you how to set that up.

Model Selection - In large projects, you don't want to plan every model when you changed one. Selectors let you target models by name, tag, wildcard, or graph operator (+model for upstream, model+ for downstream).

Data Quality - Three layers of protection: audits block invalid data during model runs, checks monitor quality trends without blocking, and tests validate transformations locally. This guide shows how to wire all three together.

Semantics¶

Transpiling Semantics - vulcan transpile converts semantic queries into executable SQL. Use it to preview, debug, and validate your semantic logic before it hits the database.

Semantic Query Lifecycle - How a semantic query travels through Vulcan: from POST request, to transpilation, cache check, worker execution, and result download. Covers the async model and all six stages.

APIs¶

Vulcan API Guide - Query your semantic layer from applications, dashboards, or BI tools. Covers the REST API (async submit, poll, download), GraphQL, and MySQL wire protocol, along with authentication, the vulcan api command, and production deployment config.

Deployment¶

Deployment Steps - Deploy your Vulcan project to a DataOS environment. Covers CLI setup, depot and compute configuration, resource manifests, and verification steps.