Google Cloud Previews Datastream Integration With Knowledge Catalog
The preview lets teams discover and govern Datastream's streams and connection profiles alongside their other cataloged data assets.
Google Cloud announced a public preview integrating Datastream, its change-data-capture and replication service, with Knowledge Catalog (formerly Dataplex), its metadata management platform. Google describes the integration as the first step toward a single, centralized view of every Datastream asset a team operates, rather than having to inspect streams and connection profiles through Datastream's own console in isolation.
Once enabled, the integration automatically discovers and synchronizes metadata for three kinds of Datastream resources: streams, including their name, location, current state, backfill strategy, and source and destination details; connection profiles, including type, connectivity option, and hostname or port; and private connectivity configurations, including network details and connection state. Synchronization runs in both live and batch modes, depending on how current a team needs the catalog to be.
Because the metadata lands in the same Knowledge Catalog that already indexes assets like BigQuery tables, teams can search for a Datastream resource using the same interface — including keyword queries like `system=Datastream AND type=Stream` or natural-language search — that they already use to find other cataloged data. Google positions this as a governance and discoverability improvement first: knowing what streams exist, where they point, and who owns them becomes a catalog query rather than a manual audit across the Datastream console.
Centralizing this kind of metadata addresses a common operational problem at larger organizations: as the number of active data pipelines grows, knowing which team owns a given stream, what it feeds into, and whether a connection profile is still in use becomes difficult to track through a single service's console alone. Folding Datastream into the same catalog used for BigQuery and other assets means that question can be answered with one search instead of cross-referencing multiple tools.
The feature is explicitly a preview and subject to Google Cloud's Pre-GA Offerings Terms, meaning the interface and behavior described here may change before general availability. Google's own documentation is the definitive, continuously updated source for how the integration works and what metadata it currently supports.
Source: Google Cloud Documentation — Datastream × Knowledge Catalog