Google Dataplex is GCP's unified data governance and management service - a "data fabric" that brings catalog, lineage, data quality, profiling, and access governance together over data distributed across BigQuery, Cloud Storage lakes, and beyond. It is where Google's governance capabilities (including the former Data Catalog) have converged into one platform.
Govern distributed data as a fabric
Dataplex's premise is that data lives in many places - multiple BigQuery datasets, several Cloud Storage buckets - and should be governed consistently without physically centralizing it. You organize assets into logical lakes and zones (e.g., raw vs curated) and apply governance, discovery, and quality across them uniformly, regardless of where the bytes sit.
What it unifies
- Catalog and discovery - search, tagging, and business metadata (the Data Catalog capability).
- Data lineage - automatic lineage for BigQuery and pipelines.
- Data quality and profiling - auto-profiling and configurable quality scans with scores.
- Governance and security - consistent access policies and classifications across the fabric.
Where it fits
- GCP organizations wanting one governance plane over BigQuery + lake data rather than separate tools.
- Teams needing built-in lineage and quality alongside the catalog, native to GCP.
- Logical organization (lakes/zones) of distributed data with uniform policy.
How it compares
Dataplex is Google's counterpart to AWS (Glue Catalog + Lake Formation) and Azure (Microsoft Purview) - a cloud-native, unified governance offering. Against third-party catalogs like Collibra, Alation, or open-source DataHub/OpenMetadata, Dataplex trades cross-cloud neutrality for deep, native GCP integration and a single bill. On GCP it's the default governance path; multi-cloud estates sometimes layer a neutral catalog on top.