Trigger

Processes all available data, possibly in multiple micro-batches, then stops.

Continuous processing with the given checkpoint interval (e.g. "1 second").

Continuous processing with the given checkpoint interval in milliseconds.

Continuous processing with the given checkpoint interval (in timeUnit).

Continuous processing with the given checkpoint interval.

Runs a micro-batch every interval of processing time (e.g. "10 seconds").

Runs a micro-batch every intervalMs milliseconds of processing time.

Runs a micro-batch every interval (in timeUnit) of processing time.

Runs a micro-batch every interval of processing time.

Real-time mode (Spark 4.2+) with batches of the given duration (e.g. "10 seconds"): records are processed as they arrive, for sources and sinks that support it.

Real-time mode with batches of batchDurationMs milliseconds.

Real-time mode with batches of the given duration (in timeUnit).

Real-time mode with batches of the given duration.

Real-time mode with the default batch duration of 5 minutes.

Processes all available data in one micro-batch, then stops.