Skip to main content
Version: Latest-3.2

CREATE PIPE

Description​

Creates a new pipe for defining the INSERT INTO SELECT FROM FILES statement used by the system to load data from a specified source data file to a destination table. This command is supported from v3.2 onwards.

Syntax​

CREATE [OR REPLACE] PIPE [db_name.]<pipe_name> 
[PROPERTIES ("<key>" = "<value>"[, "<key> = <value>" ...])]
AS <INSERT_SQL>

StarRocks supports CREATE [OR REPLACE] PIPE from v3.2.3 onwards. When you use CREATE [OR REPLACE] PIPE to create a pipe and the pipe name specified in pipe_name is the same as the name of an existing pipe in the current database, the existing database is replaced by the new pipe.

Parameters​

db_name​

The unique name of the database to which the pipe belongs.

NOTICE

Each pipe belongs to a specific database. If you drop the database to which a pipe belongs, the pipe is deleted along with the database, and cannot be recovered even if the database is recovered.

pipe_name​

The name of the pipe. The pipe name must be unique within the database in which the pipe is created. For the naming conventions, see System limits.

INSERT_SQL​

The INSERT INTO SELECT FROM FILES statement that is used to load data from the specified source data file to the destination table.

For more information about the FILES() table function, see FILES.

PROPERTIES​

A set of optional parameters that specify how to execute the pipe. Format: "key" = "value".

PropertyDefault valueDescription
AUTO_INGESTTRUEWhether to enable automatic incremental data loads. Valid values: TRUE and FALSE. If you set this parameter to TRUE, automatic incremental data loads are enabled. If you set this parameter to FALSE, the system loads only the source data file content specified at job creation and subsequent new or updated file content will not be loaded. For a bulk load, you can set this parameter to FALSE.
POLL_INTERVAL300 secondThe polling interval for automatic incremental data loads.
BATCH_SIZE1GBThe size of data to be loaded as a batch. If you do not include a unit in the parameter value, the default unit byte is used.
BATCH_FILES256The number of source data files to be loaded as a batch.

Examples​

Create a pipe named user_behavior_replica in the current database to load the data of the sample dataset s3://starrocks-examples/user_behavior_ten_million_rows.parquet to the user_behavior_replica table:

CREATE PIPE user_behavior_replica
PROPERTIES
(
    "AUTO_INGEST" = "TRUE"
)
AS
INSERT INTO user_behavior_replica
SELECT * FROM FILES
(
    "path" = "s3://starrocks-examples/user_behavior_ten_million_rows.parquet",
    "format" = "parquet",
    "aws.s3.region" = "us-east-1",
    "aws.s3.access_key" = "AAAAAAAAAAAAAAAAAAAA",
    "aws.s3.secret_key" = "BBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB"
);

NOTE

Substitute your credentials for AAA and BBB in the above command. Any valid aws.s3.access_key and aws.s3.secret_key can be used, as the object is readable by any AWS authenticated user.

This example uses the IAM user-based authentication method and a Parquet file that has the same schema as the StarRocks table. For more information about the other authentication methods and the CREATE PIPE usage, see Authenticate to AWS resources and FILES.

References​