Skip to main content
This tutorial is based on the RisingWave, Lakekeeper, Iceberg, and DuckDB demo from our awesome-stream-processing repository. For more details, please refer to the original demo.

​
Prerequisites

Before you start, ensure you have the following tools installed:
  • Docker and Docker Compose: For running the services.
  • psql: The PostgreSQL interactive terminal for connecting to RisingWave.

​
Start the stack

First, clone the demo repository and start the services using Docker Compose. 
git clone https://github.com/risingwavelabs/awesome-stream-processing.git
cd awesome-stream-processing/07-iceberg-demos/risingwave_lakekeeper_iceberg_duckdb

# Launch demo stack
docker compose up -d
This will start RisingWave, Lakekeeper, and MinIO.

​
Connect to RisingWave and stream to Iceberg

Connect to RisingWave using psql: 
psql -h localhost -p 4566 -d dev -U root
Create a REST catalog connection to Lakekeeper and MinIO: 
CREATE CONNECTION lakekeeper_catalog_conn
WITH (
    type = 'iceberg',
    catalog.type = 'rest',
    catalog.uri = 'http://lakekeeper:8181/catalog/',
    warehouse.path = 'risingwave-warehouse',
    s3.access.key = 'hummockadmin',
    s3.secret.key = 'hummockadmin',
    s3.path.style.access = 'true',
    s3.endpoint = 'http://minio-0:9301',
    s3.region = 'us-east-1'
);
This command registers the Lakekeeper REST catalog and the MinIO object storage as a connection in RisingWave, allowing it to manage Iceberg tables. Set it as the default Iceberg connection: 
SET iceberg_engine_connection = 'public.lakekeeper_catalog_conn';
Create an Iceberg-native table in RisingWave: 
CREATE TABLE customer_profile (
  customer_id INT PRIMARY KEY,
  full_name   VARCHAR,
  email       VARCHAR,
  phone       VARCHAR,
  status      VARCHAR,
  city        VARCHAR,
  created_at  TIMESTAMPTZ,
  updated_at  TIMESTAMPTZ
)
WITH (commit_checkpoint_interval = 1)
ENGINE = iceberg;
You’ve now created an Iceberg-native table directly in RisingWave. Any data streamed into this table will be committed to the Iceberg format in MinIO. Insert some data into the table: 
INSERT INTO customer_profile (
  customer_id, full_name, email, phone, status, city, created_at, updated_at)
VALUES
  (1, 'Alex Johnson', 'alex.johnson@example.com', '+1-212-555-0101', 'active', 'New York','2025-08-15 09:12:00-04','2025-08-18 14:30:00-04'),
  (2, 'Maria Garcia', 'maria.garcia@example.com', '+1-415-555-0102', 'active', 'San Francisco','2025-08-16 08:05:00-07','2025-08-19 10:45:00-07'),
  (3, 'Ethan Brown',  'ethan.brown@example.com',  '+1-312-555-0103', 'suspended','Chicago','2025-08-17 11:10:00-05','2025-08-17 11:10:00-05');
Verify that the data has been inserted: 
SELECT * FROM customer_profile;
At this point, the data is not only available for querying within RisingWave but is also persisted in the Iceberg format in your MinIO storage.

​
Query the Iceberg table from DuckDB

Now, let’s query the customer_profile table from DuckDB. First, install the DuckDB CLI: 
curl https://install.duckdb.org | sh
For DuckDB to reach MinIO, map minio-0 to 127.0.0.1 in your /etc/hosts file: 
echo "127.0.0.1 minio-0" | sudo tee -a /etc/hosts
Launch DuckDB: 
~/.duckdb/cli/latest/duckdb
Install the required extensions: 
INSTALL aws;
INSTALL httpfs;
INSTALL iceberg;
Configure DuckDB to connect to MinIO: 
SET s3_region = 'us-east-1';
SET s3_endpoint = 'http://minio-0:9301';
SET s3_access_key_id = 'hummockadmin';
SET s3_secret_access_key = 'hummockadmin';
SET s3_url_style = 'path';
SET s3_use_ssl = false;
Attach the Lakekeeper REST catalog and query the table: 
ATTACH 'risingwave-warehouse' AS lakekeeper_catalog (
  TYPE ICEBERG,
  ENDPOINT 'http://127.0.0.1:8181/catalog/',
  AUTHORIZATION_TYPE 'none'
);

SELECT * FROM lakekeeper_catalog.public.customer_profile;
You are now querying the exact same Iceberg table from a different query engine, without any data duplication. This demonstrates the open and interoperable nature of the streaming lakehouse you’ve just built with RisingWave!