Like the Aggregate node, the Rollup node can perform functional operations against each row of data produced by linked data sources. Unlike the Aggregate node, the Rollup node performs operations against several rows of data. The number of rows used to collect function argument data is determined by the rollup interval size.

You can use the Rollup Editor to create the output schema of the graph node. You create the schema by adding field names to the schema and assigning functions to each field.

When you click the Add button , an empty text field and drop down list display in the editor. To complete the addition of a field, first type a unique name in the text field for the schema field and then select an input field from the drop down list. Each field in the drop down list is a field from one of the incoming schemas.

The Rollup Editor includes an Auto button in the top right corner of the editor. Clicking this button will populate the rollup schema with fields from each incoming schema, automatically supplying unique field names.

Each function takes a set of input source schema fields as parameter values. You can use the Function Editor to select parameters to apply. There is no limit to the number of parameters that may be selected for a function, with the exception of the Pass Through function, which passes through a single value without altering the value, making the value available for the next consuming node in the graph. The Rollup interval is an absolute time-based interval that assumes the key field of the incoming data schema. You can use the Rollup Interval option list to select a rollup interval. The interval selector provides several predefined settings and also provides a custom setting the includes a custom time editor.