Transition from Dataplex Universal Catalog to Knowledge Catalog

To unify data governance with generative AI capabilities, Dataplex Universal Catalog has transitioned to Knowledge Catalog. This evolution shifts the product focus from a passive metadata registry to an active, AI-powered context graph within Google's Agentic Data Cloud. The Agentic Data Cloud is a framework that unifies databases, data warehouses, and AI tools. It builds a semantic context engine that lets AI agents take autonomous actions.

Existing Dataplex Universal Catalog configurations, APIs, and deployments remain operational. Core features like data discovery, data lineage, data quality scans, and business glossaries are unchanged and supported. Existing metadata, aspects, and configurations transition to the Knowledge Catalog experience without manual migration, data movement, or downtime.

API and client library compatibility

The update to Knowledge Catalog preserves all existing API endpoints, client libraries, and Google Cloud CLI commands:

Key differences with conventional data catalogs

Because organizations accelerate generative AI adoption, AI agents require deep business context to prevent failures and errors. Knowledge Catalog bridges the gap between enterprise data governance and AI agent workflows. It provides the contextual framework that agents need to retrieve accurate, grounded responses.

Unlike conventional passive catalogs, Knowledge Catalog automatically curates metadata, business relationships, and operational rules into a unified context graph. It introduces active, AI-centric features, including automatic context curation using unstructured data insights, verified example queries, and Model Context Protocol (MCP) integrations.

What's next

Overview

Learn about the core capabilities and architecture of Knowledge Catalog.

Concept

Discover how active data context powers generative AI grounding and eliminates hallucinations.

Tutorial

Get hands-on by creating business glossaries and custom aspect types in your project.

Reference

Explore REST and RPC APIs, client libraries, and gcloud CLI commands for automation.