User Guide

Back Next

Window Transformation Window Properties

A Window transformation has different window types that allow you to accumulate data groups at different time intervals.

Configure the following window properties for a Window transformation:

Window Type: The type of window transformation you want to create. You can choose tumbling or sliding.
Window Size: The window size defines the time interval for which data is accumulated as a data group. The window size should be a multiple of the batch interval. Specify the window size as a value in units of time or as a parameter of type TimeDuration.
Sliding Interval: The slide interval defines the time interval after which the accumulated data group is processed. Specify the slide interval as a value in units of time or as a parameter of type TimeDuration. Specify the sliding interval if you create a sliding window. By default, the window size and sliding interval are same for tumbling windows.
Watermark Delay: The watermark delay defines threshold time for a delayed event to be accumulated into a data group. Watermark delay is a threshold where you can specify the duration at which late arriving data can be grouped and processed.; If an event data arrives within the threshold time, the data is processed, and the data is accumulated into the corresponding data group. You can specify the watermark delay as a value in units of time or as a parameter of type TimeDuration in the window properties.

The following image shows sample window transformation properties: