You can also specify the bucket name with the folder path. Use an S3 bucket in the same region as your Amazon Redshift cluster.

The mapping performance improves when the Data Integration Service compresses the staging files.

Specify a directory on the machine that hosts the Data Integration Service.

Add options to the Unload command to write data from an Amazon Redshift object to an S3 bucket. You can add the following options:

Specify a directory on the machine that hosts the Data Integration Service.

Do not use

PARALLEL OFF

if you run a mapping on the Blaze engine.

Select one of the following values:

None
SSE-S3
SSE-KMS
CSE-SMK

For more information, see Data Encryption in Amazon Redshift Sources.

When the file size of an Amazon S3 object is greater than 8 MB, you can choose to download the object in multiple parts in parallel.

When you download large Amazon S3 objects, the large objects are broken into multiple parts. Default is 5 MB.

The pre-SQL commands to run a query before you read data from Amazon Redshift. You can also use the UNLOAD or COPY command. The command you specify here is processed as a plain text.

The post-SQL commands to run a query after you read data to Amazon Redshift. You can also use the UNLOAD or COPY command. The command you specify here is processed as a plain text.

The Data Integration Service includes a

SELECT DISTINCT

statement in the query if you choose this option. Amazon Redshift ignores trailing spaces. Therefore, the Data Integration Service might extract fewer rows than expected.

Enclose column names in double quotes. The SQL query is case sensitive. Specify an SQL statement supported by the Amazon Redshift database.

When you specify the columns in the SQL query, ensure that the column name matches the source column name that are mapped in the mapping.