Get the latest tech news

Crunchy Data Warehouse: Postgres with Iceberg for High Performance Analytics

We are excited to release Crunchy Data Warehouse, a modern data warehouse for Postgres. Crunchy Data Warehouse combines Postgres with Iceberg, Parquet, and data lake formats for fast analytics queries and cost efficient storage.

The versatility, reliability, performance, and extensibility of PostgreSQL make it the perfect tool for a large variety of operational workloads. The one area in which PostgreSQL has historically been lacking is analytics, which involves queries that summarize, filter, or transform large amounts of data. USING iceberg AS SELECT ..• Unlogged/temp tables• Load data directly from CSV/JSON/Parquet/GDAL• MERGE• Schema changes• INSERT .. ON CONFLICT• Renames of columns, tables, schemas, databases• TABLESAMPLE• Views• SELECT .. FOR UPDATE• Materialized views• Row-level security• Modifying CTEs• System columns• Nested data types (arrays, composite types, maps)• Logical replication• User-defined types• Materialized views stored in Iceberg• User-defined functions• Stored procedures• Prepared statements• Subtransactions (savepoints, exceptions)• Triggers• Check constraints• Sequences• Generated columns• EXPLAIN• TRUNCATE• Table-, column-, and view-level access controls• PostGIS geometry columns• Joins between Iceberg and heap tables• Prepared transactions• Export query results to CSV/JSON/Parquet file in S3• Crunchy Bridge managed service featuresAs an engineer, one of my favorite Crunchy Data Warehouse features is the fact that transactions that write to multiple Iceberg tables and heap tables are singular ACID transactions.

Get the Android app