Google Cloud Knowledge Catalog

Google Cloud Knowledge Catalog data catalog

Knowledge Catalog harvests BigQuery and Google Cloud metadata automatically, with a glossary, data products and imported dbt semantics.

Compare: data catalog tools

• Automated Metadata Harvesting: Automatically ingests metadata from BigQuery, Vertex AI, Spanner, Bigtable, Pub/Sub, Cloud SQL and Looker, plus scheduled connectors for SQL Server, PostgreSQL, Oracle and MySQL (Preview, 2026) (docs).

• Search & Discovery: Keyword and faceted search across catalog entries, scoped to resources the user can access (docs).

• AI-Powered Search: Natural language search went GA on 2025-12-08, and Knowledge Studio (Preview, 2026-11-02) adds natural-language exploration of a context graph (docs).

• Custom Fields & Templates: Custom aspect types define structured metadata templates that attach to any entry (docs).

• Business Glossary: Business glossaries hold terms and categories that link to entries and columns (docs).

• Usage & Popularity Statistics: No per-asset usage or popularity view is documented; system aspects can be populated from query logs but usage metrics aren't named (docs).

• Catalog Usage Analytics: No catalog adoption or documentation-coverage dashboard is documented (docs).

• Metadata Propagation: Aspects attach to individual entries; no automatic cascade to children or downstream assets is documented (docs).

• Asset Certification: No built-in certification or endorsement status; the docs only show a user-defined CertifiedDataProduct entry type as an example (docs).

Google Cloud Knowledge Catalog data lineage

Knowledge Catalog captures table and column lineage for Google Cloud services and accepts OpenLineage events.

Compare: data lineage tools

• Table-Level Lineage: Data lineage is recorded automatically for BigQuery jobs, Managed Airflow, Vertex AI Pipelines and Data Fusion once the Data Lineage API is enabled (docs).

• Column-Level Lineage: Column-level lineage is captured for BigQuery and Lakehouse Iceberg tables, and for Dataproc/Spark (GA 2026-05-15) (docs).

• End-to-End Lineage: Looker lineage went GA on 2026-10-26, showing flow from BigQuery sources through the Looker semantic layer to dashboards and Looks (docs).

• Impact Analysis: Lineage graphs and the lineage MCP server (Preview, 2026-05-27) are documented for analyzing downstream impact of changes (docs).

• OpenLineage Support: OpenLineage run events can be imported through the Data Lineage API (ProcessOpenLineageRunEvent), though custom events don't carry column-level lineage (docs).

Google Cloud Knowledge Catalog data stewardship

Compare: data stewardship tools

• Review & Approval Workflows: Governance workflows (Preview, 2026-07-24) add a request-review mechanism for data product access, and data product owners approve or reject requests (GA 2026-05-25) (docs).

• Governance Campaigns: Bulk metadata import jobs exist, but no steward-driven bulk campaign feature is documented (docs).

• Metadata Automation: Metadata change feeds publish catalog changes to Pub/Sub for event-driven workflows, but rule-based actions must be built by the customer (docs).

• Critical Data Elements: No critical data element registry is documented (docs).

• Governance Scorecards: No questionnaire-based governance maturity scoring is documented (docs).

Google Cloud Knowledge Catalog data classification

Compare: data classification tools

• Sensitive Data Discovery: Sensitive Data Protection discovery profiles BigQuery and other data and writes sensitivity findings to Knowledge Catalog entries as aspects (docs).

• Data Policy: Security policies are applied and propagated across lakes and zones at no charge, and BigQuery column-level security uses policy tags (pricing).

• Data Masking: BigQuery dynamic data masking applies masking rules through policy tags on columns (docs).

• Native Policy Enforcement: Policies are enforced in Google Cloud services (BigQuery, Cloud Storage) that are part of the same platform, which this row excludes (docs).

Google Cloud Knowledge Catalog AI features

Gemini enriches metadata and answers questions, and Knowledge Catalog serves context to agents.

• AI Chat Assistant: Knowledge Studio (Preview, 2026-11-02) lets users explore assets and relationships in natural language, and a discovery agent handles complex natural-language queries (2026-04-17) (docs).

• AI Documentation Generation: Data insights uses Gemini to generate table descriptions, summaries and sample queries (docs).

• Agentic Metadata Enrichment: Knowledge Studio enrichment tasks (Preview) generate and curate metadata from schemas, query logs and enterprise knowledge signals; billing starts at GA (docs).

Google Cloud Knowledge Catalog data compliance

• Audit Trail: Knowledge Catalog writes admin activity and data access entries to Cloud Audit Logs (docs).

• Data Activity Monitoring: Data access is logged in Cloud Audit Logs, but Knowledge Catalog documents no access-pattern analysis or anomaly detection (docs).

• Compliance Reporting: No compliance evidence packs or regulatory reports are documented (docs).

• Regulatory Framework Mapping: No mapping of assets or policies to GDPR, HIPAA or other regulations is documented (docs).

• ROPA: No GDPR Article 30 processing register is documented (docs).

• Data Retention Policies: No retention or deletion policy management is documented in Knowledge Catalog (docs).

• Data Subject Rights: No DSAR or right-to-be-forgotten workflow is documented (docs).

Google Cloud Knowledge Catalog collaboration

• Alert Notifications: Data quality scans can send email notifications, and metadata change feeds go to Pub/Sub; no native Slack or Teams alerts are documented (docs).

• Discussions & Annotations: No comments or threaded discussions on entries are documented (docs).

• Ticketing Integrations: No Jira or ServiceNow integration is documented (docs).

Google Cloud Knowledge Catalog data marketplace

Compare: data marketplace tools

• Self-Service Access Requests: Data product consumers browse published products, submit access requests and track their status (GA 2026-05-25) (docs).

• Data Products: Data products package assets with owners, contracts and access groups and went GA on 2026-05-25; Looker and Iceberg REST assets are supported (docs).

• Data Mesh Support: Lakes, zones and data domains (Preview, 2026-09-07) organize resources by domain, and Google positions Knowledge Catalog for building a data mesh (docs).

• Subscription Management: The data products docs don't say whether consumers are notified of changes to a product or its contract (docs).

Google Cloud Knowledge Catalog data quality

• SQL Table Tests: Auto data quality runs predefined and custom SQL rules on BigQuery tables, with reusable rule templates since 2026-04-17 (docs).

• Data Health Scores: Data quality scans produce a quality score per scan and per dimension that can be published to the catalog entry (docs).

• Data Profiling: Data profile scans compute column statistics, with a lightweight low-latency mode (Preview, 2026-04-10) and Gemini profiling of unstructured data (Preview) (docs).

• Anomaly Detection: Auto data quality is rule-based; ML-based anomaly detection on values or volumes isn't documented in Knowledge Catalog (docs).

• Data Contracts: Data products carry a contract stating refresh cadence and quality thresholds, but enforcement or alerting on breaches isn't documented (docs).

Google Cloud Knowledge Catalog AI governance

• AI Asset Inventory: Vertex AI models, datasets, feature groups, feature views and online stores are ingested into the catalog automatically (docs).

• AI Lineage: Lineage is enabled automatically for Vertex AI Pipelines, and Vertex AI Feature Store metadata is tracked (docs).

• AI Agent Access Governance: lookupContext (Preview, 2026-06-04) and MCP servers deliver catalog context to agents, but enforcement of policy at agent query time beyond IAM isn't documented (docs).

• AI Regulatory Compliance: No EU AI Act, NIST AI RMF or ISO 42001 templates or evidence workflows are documented (docs).

• AI Model Health Score: No composite trust or health score per model is documented (docs).

Google Cloud Knowledge Catalog workflows

• REST / GraphQL API: The Dataplex REST API covers entries, aspects, glossaries, scans and lineage (docs).

• MCP Server: Remote MCP servers expose data products (Preview, 2026-05-25) and data lineage (Preview, 2026-05-27) to AI agents (docs).

• dbt Integration: Imports dbt Core, dbt Fusion and dbt Cloud metadata, including MetricFlow semantics, tests and lineage linked to BigQuery tables (GA 2026-09-24) (docs).

• Python SDK: The google-cloud-dataplex client library is pip-installable (docs).

• Webhooks: Metadata change feeds (2026-02-11) publish near real-time catalog change notifications to a Pub/Sub topic (docs).

• Browser Extension: No browser extension is documented (docs).

Google Cloud Knowledge Catalog data context layer

• Knowledge Graph: Knowledge Studio (Preview, 2026-11-02) provides an interactive context graph, and the catalog discovers relationships between assets (Preview, 2026-04-20) (docs).

• Semantic Models: MetricFlow semantics can be imported from dbt and Looker semantic-layer lineage is shown, but the catalog doesn't generate its own semantic model (docs).