Datastream for BigQuery Preview
Seamless replication from relational databases directly to BigQuery, enabling near real-time insights on operational data.
Replicate operational data with minimal latency
Seamlessly replicate data from MySQL, PostgreSQL, AlloyDB, and Oracle databases directly into BigQuery, with low latency and without impacting source performance.
Scale up and down with a serverless architecture
Eliminate operational overhead with a serverless approach that scales automatically with no infrastructure for you to manage.
Get up and running in minutes
A simplified setup experience allows you to start replicating data from your operational databases to BigQuery in just a few steps.
Datastream reads change events (inserts, updates, and deletes) from source databases and writes them in BigQuery tables in near real time. This enables you to enrich existing BigQuery data warehouses and ML models with transactional data, such as retail purchases, to build a more complete end-to-end picture of data. Datastream will backfill historical data, continuously replicate new changes as they happen, and seamlessly handle schema changes.
Datastream for BigQuery
Datastream and Dataflow
Datastream and Data Fusion
Easiest option for replicating operational data to BigQuery
Serverless architecture that automatically scales up and down
Single interface for end-to-end visibility and monitoring of replication pipelines
Customizable solution with additional flexibility
Pre-built templates supported by Google for a range of destinations
Integration of additional features such as data quality and data masking
Simple interface for ETL developers and data analysts
Identification of potential issues and gaps in replication in advance
Near real-time insights into replication performance