Memory Group By

The Memory Group By step groups rows in memory from a source step. The resulting rows are grouped based on a specified field or collection of fields. A new row is generated for each group.

This step differs from the Group By step by processing all rows in memory and by handling non-sorted input.

If the number of rows you want to group is too large to fit into memory, use a combination of Sort rows and Group By.

General

Option	Definition
Step name	Specify the unique name of the Memory Group By step on the canvas. You can customize the name or leave it as the default.
Always give back a result row	Select to return a result row even when there is no input row. If no input rows exist, this option returns a count of 0.

Memory Group By dialog

Group fields

Use this table to specify the fields you want to group.

• Select Get fields to add all fields from the incoming stream.

• Leave this table blank to calculate aggregate functions over the entire dataset.

Aggregates

Use this table to specify the aggregation method and the name of the resulting field.

Column	Description
Name	The name of the aggregate field.
Subject	The subject on which you want to use an aggregation method.
Type	The aggregation method. Available methods include: Sum Average (Mean) Median Percentile Minimum Maximum Number of values (N) Concatenate strings separated by , (comma) First non-null value Last non-null value First value (including null) Last value (including null) Standard deviation Concatenate strings separated by the character specified in the Value column Number of distinct values Number of rows (without field argument)
Value	The aggregate value.

Metadata injection support

This step supports metadata injection. You can use it with ETL metadata injection to pass metadata to your transformation at runtime.