APACHE ICEBERG
Real-Time CDC to Open Lakehouse
Stream database changes directly to Apache Iceberg tables. Open format with zero vendor lock-in. Query from Spark, Trino, Athena, Databricks, or any engine. ACID transactions and time travel included.
Open Table Format, Zero Lock-in
Write once, query from anywhere with full transactional guarantees
Open Table Format
No vendor lock-in. Query Iceberg tables from Spark, Trino, Presto, Dremio, Athena, and more.
ACID Transactions
Full transactional guarantees for concurrent reads and writes. No corrupt data.
Time Travel
Query historical snapshots. Roll back to any point in time for debugging or compliance.
Schema Evolution
Add, rename, or drop columns without rewriting data. Changes flow automatically.
Hidden Partitioning
Automatic partition pruning without exposing partitions in queries.
Any Cloud Storage
Write to S3, GCS, Azure Blob, MinIO, or any S3-compatible storage.
Supported Catalogs
Connect to your existing metadata catalog
AWS Glue
Native Glue Data Catalog integration
Hive Metastore
Self-hosted Hive Metastore
REST Catalog
Tabular, Nessie, or custom REST
Snowflake
Polaris Catalog integration
What You Can Build
How It Works
Streamkap captures changes from your source databases using CDC and streams them directly to Apache Iceberg tables on your cloud object storage. Query with any engine.
- 1. Capture: CDC captures changes from PostgreSQL, MySQL, MongoDB, and more
- 2. Transform: Apply SQL transforms, filtering, and masking in-flight
- 3. Write: Data written to Iceberg tables with ACID guarantees
- 4. Query: Use Spark, Trino, Athena, or any Iceberg-compatible engine
Query with Any Engine
Iceberg's open format works with every major query engine
Open lakehouse with real-time CDC. Zero lock-in.
Query from any engine. Write to any cloud storage. Set up in minutes.