Vegalake documentation

Vegalake is a cloud data and AI platform made of three connected products. VegaDB runs distributed analytical SQL. VegaFlow integrates systems and operates data and AI workflows. VegaGraph links their datasets and runs to owners, policies, services and business definitions.

This documentation covers connections, catalogs, warehouses, workflows, graph objects, lifecycle states, limits and failure behavior.

New to the workspace? Start with organizations and workspaces, access management and secret vaults. Use API fundamentals when connecting automation.

VegaDB

Connect with a PostgreSQL client, create a warehouse and run distributed analytical SQL.

VegaFlow

Create managed connections, QuickFlows and durable data or ML pipelines.

VegaGraph

Model entities, aspects and relationships; search and traverse them with context.

How the products fit

VegaFlow normally begins the path. It reads from operational databases, object stores or event systems and writes into the lakehouse. VegaDB turns that data into something analysts and applications can query. VegaGraph records what a table means, who owns it, what produced it and which KPI will change if it breaks.

You can use any product alone. When used together, the products share identity and lineage across datasets, runs and graph entities.

Sources → VegaFlow → VegaDB → dashboards / applications
             │          │
             └──── VegaGraph ──── ownership, lineage, policy, evidence

Pick a starting point

Start with Connect to VegaDB. It covers the PostgreSQL endpoint, TLS, supported clients and the session model.

Cloud service

The docs describe Vegalake Cloud. Host names, compute placement, credentials and storage integrations are managed through your organization and workspace.