Joining Heterogeneous Databases is a reality, not a Myth
summary
Demonstrates that PostgreSQL Foreign Data Wrappers let one actually join tables across Postgres, ClickHouse, MySQL and even MongoDB by pushing work to the remote side rather than shipping rows. Uses postgres_fdw plus JDBC/ODBC wrappers and a ClickHouse FDW, shows query plans with join pushdown and aggregate pushdown, compares network bytes and latency on real queries, and builds a working prototype where remote execution cuts data transfer and speeds up multi-source analytic joins.
AI-written summary · about
chapters (8)
- 0:00Introduction to joining databases
- 3:48Application architecture challenges
- 8:23Database access methods
- 14:06Testing and server access
- 20:41Table naming and mapping issues
- 26:11Remote system integration
- 30:45Data selection and editing
- 33:27Q&A and conclusion
related talks
39mPostgreSQL: Joining Heterogeneous Databases is a reality, not a Myth
50mBeyond Foreign Data Wrappers: building a data warehouse from heterogeneous external data sources
51mUsing Postgres FDW with a sharded PostgreSQL Database
54mFederating Queries Using postgres_fdw
52mSharding PostgreSQL using Foreign Data Wrappers
48mData Horizons With Postgres
46mWriting a Foreign Data Wrapper
30mThe database trends that are transforming your database infrastructure forever