ICEBERG_BUCKET

Hashes a value using MurmurHash3 and assigns it to a bucket number, according to the Iceberg partition transforms specification.

Syntax

ICEBERG_BUCKET(<value>, <num_buckets>)

Parameters

Parameter	Description	Supported input types
`<value>`	The value to hash into a bucket.	`INTEGER`, `BIGINT`, `TEXT`, `BYTEA`, `DATE`, `TIMESTAMP`, `TIMESTAMPTZ`
`<num_buckets>`	The number of buckets. Must be a positive constant.	`BIGINT`

Return type

INTEGER Returns a bucket number from 0 to num_buckets - 1.

Remarks

The ICEBERG_BUCKET function can be used in the PARTITION BY clause of CREATE ICEBERG TABLE commands.

CREATE ICEBERG TABLE events
  PARTITION BY (iceberg_bucket(user_id, 16))
  AS SELECT * FROM source_events
WITH LOCATION = my_iceberg_location;

The bucket number is calculated by applying MurmurHash3 to the value, then taking the modulo with the number of buckets. The hash calculation follows the Iceberg 32-bit hash requirements. The <num_buckets> parameter must be a positive constant value. Using zero, negative values, or non-constant expressions results in an error.

Example

SELECT iceberg_bucket(34, 1024) AS bucket_partition;

bucket_partition int
339

Rows: 1Execution time: 5.11ms

ICEBERG_HOUR

ICEBERG_TRUNCATE

⌘I

Overview

Performance and Observability

Security

Self-Managed

Managed service

Guides

SQL reference

Release notes

API reference

Legal

Syntax

Parameters

Return type

Remarks

Example

​Syntax

​Parameters

​Return type

​Remarks

​Example

Syntax

Parameters

Return type

Remarks

Example