Airbyte's Change Data Capture (CDC) feature could potentially improve your Postgres to BigQuery latency. CDC allows for more granular data change tracking and is one of the most efficient ways to ingest data, especially when dealing with large databases or the need to track delete operations in the source database [[source](
https://airbyte.com/tutorials/postgresql-change-data-capture-elt)]. However, it's important to note that CDC has certain requirements and limitations. For instance, incremental sync is only supported for tables with primary keys, data must be in tables and not views, and the modifications you want to capture must be made using `DELETE`/`INSERT`/`UPDATE` [[source](
https://docs.airbyte.com/integrations/sources/postgres/postgres-troubleshooting)]. As for parallelizing the different tables, Airbyte currently does not support this feature. Each Airbyte database source using CDC replication can only be used with a single Airbyte destination due to how Postgres CDC is implemented [[source](
https://docs.airbyte.com/integrations/sources/postgres/postgres-troubleshooting)]. You do not need to create a connection per table. When setting up a connection in Airbyte, you can select the tables you want to sync from your Postgres source to your BigQuery destination [[source](
https://airbyte.com/tutorials/postgresql-change-data-capture-elt)]. Before you proceed with CDC, it's recommended to confirm that Airbyte can connect to your Postgres database. For CDC, you must connect to primary/master databases. Pointing the connector configuration to replica database hosts for CDC will lead to failures [[source](
https://docs.airbyte.com/integrations/sources/postgres)]. Remember to carefully consider these factors and your specific use case before deciding to implement CDC.