Sink data from RisingWave to Apache Iceberg

In RisingWave, you can stream data into Iceberg tables with the built-in Iceberg sink connector.

Apache Iceberg is a table format designed to support huge tables. For more information, see Apache Iceberg.

​

Prerequisites

• Ensure you already have an Iceberg table that you can sink data to. For additional guidance on creating a table and setting up Iceberg, refer to this quickstart guide on creating an Iceberg table.
• Ensure you have an upstream materialized view or source that you can sink data from.

​

Syntax

CREATE SINK [ IF NOT EXISTS ] sink_name
[FROM sink_from | AS select_query]
WITH (
   connector='iceberg',
   connector_parameter = 'value', ...
);

​

Parameters

​

Basic parameters

​

Use GCS as object storage for Iceberg

RisingWave supports creating Iceberg sinks on GCS with catalog types storage or rest. To enable it, configure the following specific parameters:

CREATE SINK sink_t FROM t WITH (
    connector = 'iceberg',
    type = 'append-only',
    force_append_only = 'true',
    database.name = 'public',
    table.name = 't',
    catalog.name = 'demo',
    catalog.type = 'rest',
    catalog.uri = 'http://127.0.0.1:8181',
    warehouse.path = 'gs://bucket/path',
    gcs.credential = 'xxxxxxxx'
);

For more details on sinking data to GCS, see Sink data to Google Cloud Storage.

​

Use Azure Blob as object storage for Iceberg

RisingWave supports creating Iceberg sinks on Azure Blob with catalog types storage or rest. To enable it, configure the following specific parameters:

CREATE SINK sink1 AS select * from mv1 WITH (
    connector = 'iceberg',
    type = 'append-only',
    database.name = 'demo_db',
    table.name = 'test_azblob_iceberg',
    catalog.name = 'demo',
    catalog.type = 'storage',
    warehouse.path = 'azblob://<demo>',
    azblob.account_name = 'xxx',
    azblob.account_key = 'xxx',
    azblob.endpoint_url = 'https://<account_name>.blob.core.windows.net/',
    ...
);

For more details on sinking data to Azure Blob, see Sink data to Azure Blob.

​

Use Amazon S3 Tables with the Iceberg sink

You can configure the RisingWave Iceberg sink connector to use Amazon S3 Tables as its catalog. This setup allows RisingWave to sink data into Iceberg tables managed by the AWS native S3 Tables catalog service.

To achieve this, specify the rest catalog type within your CREATE SINK statement and include the necessary parameters for SigV4 authentication against the S3 Tables REST API.

Required REST Catalog Parameters for S3 Tables:

CREATE SINK my_s3_tables_sink FROM source_table
WITH (
    connector = 'iceberg',
    type = 'upsert', -- Or 'append-only'
    primary_key = 'id', -- Required for 'upsert' type

    -- Specify the S3 Tables warehouse ARN
    warehouse.path = 'arn:aws:s3tables:<your-region>:<your-account-id>:bucket/<your-bucket-name>',
    -- AWS Credentials
    s3.access.key = '<your-aws-access-key-id>',
    s3.secret.key = '<your-aws-secret-access-key>',
    s3.region = '<your-region>', -- e.g., 'us-east-1'

    -- S3 Tables REST catalog endpoint
    catalog.uri = 'https://s3tables.<your-region>.amazonaws.com/iceberg',
    -- REST catalog signing configurations
    catalog.rest.signing_region = '<your-region>', -- e.g., 'us-east-1'
    catalog.rest.sigv4_enabled = true,
    catalog.rest.signing_name = 's3tables',
    -- Specify REST catalog type
    catalog.type = 'rest',

    -- Target Iceberg table details within S3 Tables catalog
    database.name = '<your-database-name>', -- Database in S3 Tables
    table.name = '<your-table-name>',       -- Table name in S3 Tables
    create_table_if_not_exists = true      -- Optional: Create table if it doesn't exist
);

Once created, this sink will write data from source_table into the specified Iceberg table (<your-table-name>) within the <your-database-name> database, using Amazon S3 Tables to manage the table’s metadata.

​

Data type mapping

RisingWave converts RisingWave data types from/to Iceberg according to the following data type mapping table:

​

Catalog

Iceberg supports these types of catalogs:

​

Storage catalog

The Storage catalog stores all metadata in the underlying file system, such as Hadoop or S3. Currently, we only support S3 as the underlying file system.

CREATE SINK sink_demo_storage FROM t
WITH (
    connector = 'iceberg',
    type = 'append-only',
    force_append_only = true,
    s3.endpoint = 'http://minio-0:9301',
    s3.access.key = 'xxxxxxxxxx',
    s3.secret.key = 'xxxxxxxxxx',
    s3.region = 'ap-southeast-1',
    catalog.type = 'storage',
    catalog.name = 'demo',
    warehouse.path = 's3://icebergdata/demo',
    database.name = 's1',
    table.name = 't1'
);

​

REST catalog

RisingWave supports the REST catalog, which acts as a proxy to other catalogs like Hive, JDBC, and Nessie catalog. This is the recommended approach to use RisingWave with Iceberg tables.

CREATE SINK sink_demo_rest FROM t
WITH (
    connector = 'iceberg',
    type = 'append-only',
    force_append_only = true,
    s3.endpoint = 'https://s3.ap-southeast-2.amazonaws.com',
    s3.region = 'ap-southeast-2',
    s3.access.key = 'xxxx',
    s3.secret.key = 'xxxx',
    s3.path.style.access = 'true',
    catalog.type = 'rest',
    catalog.uri = 'http://localhost:8181/api/catalog',
    warehouse.path = 'quickstart_catalog',
    database.name = 'ns',
    table.name = 't1',
    catalog.credential='123456:123456',
    catalog.scope='PRINCIPAL_ROLE:ALL',
    catalog.oauth2_server_uri='xxx'
    catalog.scope='xxx',
);

​

Hive catalog

RisingWave supports the Hive catalog. You need to set catalog.type to hive to use it.

CREATE SINK sink_demo_hive FROM t
WITH (
    connector = 'iceberg',
    type = 'append-only',
    force_append_only = true,
    catalog.type = 'hive',
    catalog.uri = 'thrift://metastore:9083',
    warehouse.path = 's3://icebergdata/demo',
    s3.endpoint = 'http://minio-0:9301',
    s3.access.key = 'xxxxxxxxxx',
    s3.secret.key = 'xxxxxxxxxx',
    s3.region = 'ap-southeast-1',
    catalog.name = 'demo',
    database.name = 's1',
    table.name = 't1'
);

​

JDBC catalog

RisingWave supports the JDBC catalog.

CREATE SINK sink_demo_jdbc FROM t
WITH (
    connector = 'iceberg',
    type = 'append-only',
    force_append_only = true,
    warehouse.path = 's3://icebergdata/demo',
    s3.endpoint = 'http://minio-0:9301',
    s3.access.key = 'xxxxxxxxxx',
    s3.secret.key = 'xxxxxxxxxx',
    s3.region = 'ap-southeast-1',
    catalog.name = 'demo',
    catalog.type = 'jdbc',
    catalog.uri = 'jdbc:postgresql://postgres:5432/iceberg',
    catalog.jdbc.user = 'admin',
    catalog.jdbc.password = '123456',
    database.name = 's1',
    table.name = 't1'
);

​

Glue catalog

RisingWave supports the Glue catalog. You should use AWS S3 if you use the Glue catalog. Below are example codes for using this catalog:

CREATE SINK sink_test FROM t
WITH (
      type='upsert',
      primary_key='col',
      connector = 'iceberg',
      catalog.type = 'glue',
      catalog.name = 'test',
      warehouse.path = 's3://my-iceberg-bucket/test',
      s3.access.key = 'xxxxxxxxxx',
      s3.secret.key = 'xxxxxxxxxx',
      s3.region = 'ap-southeast-2',
      database.name='test_db',
      table.name='test_table'
  );

​

Iceberg table format

Currently, RisingWave only supports Iceberg tables in format v2.

​

Exactly-once delivery

RisingWave provides exactly-once delivery semantics for Iceberg sinks. This semantics guarantees that each data event is processed once and only once, even in the presence of failures such as retries or restarts. This level of delivery assurance is essential in scenarios where duplicate records can lead to incorrect analytics or data corruption in downstream systems.

Exactly-once delivery is achieved through a two-phase commit protocol involving a pre-commit phase and a commit phase. Iceberg’s commit operations are idempotent, which allows RisingWave to safely retry failed transactions without introducing duplicates.

By default, exactly-once semantics is disabled. To enable it for an Iceberg sink, include is_exactly_once = 'true' in the WITH clause of the sink definition. Note that enabling this option introduces additional coordination overhead due to metadata pre-commit, which may impact sink performance in high-throughput workloads.

​

Examples

This section includes several examples that you can use if you want to quickly experiment with sinking data to Iceberg.

​

Create an Iceberg table (if you do not already have one)

Set create_table_if_not_exists to true to automatically create an Iceberg table.

Alternatively, use Spark to create a table. For example, the following spark-sql command creates an Iceberg table named table under the database dev in AWS S3. The table is in an S3 bucket named my-iceberg-bucket in region ap-southeast-1 and under the path path/to/warehouse. The table has the property format-version=2, so it supports the upsert option. There should be a folder named s3://my-iceberg-bucket/path/to/warehouse/dev/table/metadata.

Note that only S3-compatible object store is supported, such as AWS S3 or MinIO.

spark-sql --packages org.apache.iceberg:iceberg-spark-runtime-3.4_2.12:1.3.1,org.apache.hadoop:hadoop-aws:3.3.2\
    --conf spark.sql.catalog.demo=org.apache.iceberg.spark.SparkCatalog \
    --conf spark.sql.catalog.demo.type=hadoop \
    --conf spark.sql.catalog.demo.warehouse=s3a://my-iceberg-bucket/path/to/warehouse \
    --conf spark.sql.catalog.demo.hadoop.fs.s3a.endpoint=https://s3.ap-southeast-1.amazonaws.com \
    --conf spark.sql.catalog.demo.hadoop.fs.s3a.path.style.access=true \
    --conf spark.sql.catalog.demo.hadoop.fs.s3a.access.key=${ACCESS_KEY} \
    --conf spark.sql.catalog.demo.hadoop.fs.s3a.secret.key=${SECRET_KEY} \
    --conf spark.sql.defaultCatalog=demo \
    --e "drop table if exists demo.dev.`table`;

CREATE TABLE demo.dev.`table`
(
  seq_id bigint,
  user_id bigint,
  user_name string
) TBLPROPERTIES ('format-version'='2')";

​

Create an upstream materialized view or source

The following query creates an append-only source. For more details on creating a source, see CREATE SOURCE .

CREATE SOURCE s1_source (
     seq_id bigint,
     user_id bigint,
     user_name varchar)
WITH (
     connector = 'datagen',
     fields.seq_id.kind = 'sequence',
     fields.seq_id.start = '1',
     fields.seq_id.end = '10000000',
     fields.user_id.kind = 'random',
     fields.user_id.min = '1',
     fields.user_id.max = '10000000',
     fields.user_name.kind = 'random',
     fields.user_name.length = '10',
     datagen.rows.per.second = '20000'
 ) FORMAT PLAIN ENCODE JSON;

Another option is to create an upsert table, which supports in-place updates. For more details on creating a table, see CREATE TABLE .

CREATE TABLE s1_table (
     seq_id bigint,
     user_id bigint,
     user_name varchar)
WITH (
     connector = 'datagen',
     fields.seq_id.kind = 'sequence',
     fields.seq_id.start = '1',
     fields.seq_id.end = '10000000',
     fields.user_id.kind = 'random',
     fields.user_id.min = '1',
     fields.user_id.max = '10000000',
     fields.user_name.kind = 'random',
     fields.user_name.length = '10',
     datagen.rows.per.second = '20000'
 ) FORMAT PLAIN ENCODE JSON;

​

Append-only sink from append-only source

If you have an append-only source and want to create an append-only sink, set type = append-only in the CREATE SINK SQL query.

CREATE SINK s1_sink FROM t1_table
WITH (
    connector = 'iceberg',
    type = 'append-only',
    warehouse.path = 's3a://my-iceberg-bucket/path/to/warehouse,
    s3.endpoint = 'https://s3.ap-southeast-1.amazonaws.com',
    s3.access.key = '${ACCESS_KEY}',
    s3.secret.key = '${SECRET_KEY},
    database.name='dev',
    table.name='table'
);

​

Append-only sink from upsert source

If you have an upsert source and want to create an append-only sink, set type = append-only and force_append_only = true. This will ignore delete messages in the upstream, and to turn upstream update messages into insert messages.

CREATE SINK s1_sink FROM s1_table
WITH (
    connector = 'iceberg',
    type = 'append-only',
    force_append_only = 'true',
    warehouse.path = 's3a://my-iceberg-bucket/path/to/warehouse,
    s3.endpoint = 'https://s3.ap-southeast-1.amazonaws.com',
    s3.access.key = '${ACCESS_KEY}',
    s3.secret.key = '${SECRET_KEY},
    database.name='dev',
    table.name='table'
);

​

Upsert sink from upsert source

In RisingWave, you can directly sink data as upserts into Iceberg tables.

CREATE SINK s1_sink FROM s1_table
WITH (
    connector = 'iceberg',
    type = 'upsert',
    warehouse.path = 's3a://my-iceberg-bucket/path/to/warehouse,
    s3.endpoint = 'https://s3.ap-southeast-1.amazonaws.com',
    s3.access.key = '${ACCESS_KEY}',
    s3.secret.key = '${SECRET_KEY},
    database.name='dev',
    table.name='table',
    primary_key='seq_id'
);