pyspark.sql.DataFrame.withWatermark

DataFrame.withWatermark(eventTime: str, delayThreshold: str) → pyspark.sql.dataframe.DataFrame[source]

Defines an event time watermark for this DataFrame. A watermark tracks a point in time before which we assume no more late data is going to arrive.

Spark will use this watermark for several purposes:
  • To know when a given time window aggregation can be finalized and thus can be emitted when using output modes that do not allow updates.

  • To minimize the amount of state that we need to keep for on-going aggregations.

The current watermark is computed by looking at the MAX(eventTime) seen across all of the partitions in the query minus a user specified delayThreshold. Due to the cost of coordinating this value across partitions, the actual watermark used is only guaranteed to be at least delayThreshold behind the actual event time. In some cases we may still process records that arrive more than delayThreshold late.

New in version 2.1.0.

Parameters
eventTimestr

the name of the column that contains the event time of the row.

delayThresholdstr

the minimum delay to wait to data to arrive late, relative to the latest record that has been processed in the form of an interval (e.g. “1 minute” or “5 hours”).

Notes

This API is evolving.

>>> from pyspark.sql.functions import timestamp_seconds
>>> sdf.select(
...    'name',
...    timestamp_seconds(sdf.time).alias('time')).withWatermark('time', '10 minutes')
DataFrame[name: string, time: timestamp]