Windowing Aggregator
Supported pipeline types:
|
The Windowing Aggregator processor performs one or more aggregations within a window of time. The Windowing Aggregator processor displays the results in Monitor mode and can write the results to event records. The processor does not update records in the data stream.
You might use aggregation results with Data Collector data alerts to generate threshold notifications. Or, you might write the events to a data store and use a third-party dashboard application to visualize the data.
When you configure the Windowing Aggregator, you specify the window type and time window to use. This information determines how the processor displays data and generates events. You also specify the time zone to use and the aggregations to perform.
When you configure an aggregation, you specify the aggregation name and display title and the aggregation function to use. You can optionally define a filter expression to process a subset of the data, and a group by expression to aggregate based on groups.
To write the aggregation results to events, enable event generation, configure the type of event records to create, and optionally specify the type of root field to use. For general information about dataflow triggers and the event framework, see Dataflow Triggers Overview.