You can create a target table with clustered columns at runtime. Clustering organizes the data based on the specified columns into optimally-sized storage blocks, which enhances the query performance.
The clustering order determines the sort order of the data within the table. To define the clustering order, enter up to four fields, each separated by a comma.
You can use the following data types in the clustering columns:
Bignumeric
Bool
Date
Datetime
INT64
Numeric
String
Timestamp
Rules and guidelines
Consider the following rules and guidelines when you configure clustered tables in the
Google BigQuery
target:
When you set the clustering order, the mapping can fail for the following issues:
The number of clustering fields exceeds the defined limit.
An incorrect name for the clustering field.
The clustering field is of the Byte or Float data type.
The related error message for the mapping failure is not logged in the session log. You can find the related error message logged in the Tomcat log.
When you use Simple connection mode in a mapping, you need to provide the clustering field name for the Record data type in the
Clustering Order
property using an underscore instead of a dot. For example, provide the field name as
_(Master_String)
instead of
.(Master.String)
. If you use a dot in the clustering field name, the mapping fails.