You can set up the Group By step to run on the Spark engine.
When Spark is processing the transformation, the rows are sorted by the group fields. The field names cannot contain spaces, dashes, or special characters. Each field name must start with a letter.
Optionally, you can use the Sort step before
the Group By step. If your existing transformations
contains a Sort step before the Group By step, it will run successfully.
Note: The Group
By and the Memory Group By steps work the
same.