nerdexam
Microsoft

DP-203 · Question #358

You have an Azure data factory named ADF1 that contains a pipeline named Pipeline1. Pipeline1 must execute every 30 minutes with a 15-minute offset. You need to create a trigger for Pipeline1. The…

The correct answer is D. tumbling window. Tumbling window triggers are purpose-built for exactly this scenario: they fire on a fixed, non-overlapping schedule with a defined offset, natively support backfilling historical data from a past start time to the present, enforce concurrency limits (you can set max concurrent…

Submitted by luis.pe· Mar 30, 2026Develop data processing

Question

You have an Azure data factory named ADF1 that contains a pipeline named Pipeline1. Pipeline1 must execute every 30 minutes with a 15-minute offset. You need to create a trigger for Pipeline1. The trigger must meet the following requirements:

  • Backfill data from the beginning of the day to the current time.
  • If Pipeline1 fails, ensure that the pipeline can re-execute within

the same 30-minute period.

  • Ensure that only one concurrent pipeline execution can occur.
  • Minimize development and configuration effort.

Which type of trigger should you create?

Options

  • Aschedule
  • Bevent-based
  • Cmanual
  • Dtumbling window

How the community answered

(47 responses)
  • A
    4% (2)
  • B
    13% (6)
  • C
    9% (4)
  • D
    74% (35)

Explanation

Tumbling window triggers are purpose-built for exactly this scenario: they fire on a fixed, non-overlapping schedule with a defined offset, natively support backfilling historical data from a past start time to the present, enforce concurrency limits (you can set max concurrent runs to 1), and have built-in retry policies so a failed run can re-execute within the same window period - all configured in a single trigger definition.

Schedule triggers (A) lack native backfill support and do not guarantee single-window isolation; they also don't retry within the same time window. Event-based triggers (B) fire in response to storage events (like a file arriving), not on a time interval - completely wrong pattern here. Manual triggers (C) require human intervention each time, which obviously can't meet a 30-minute recurring requirement with backfill.

Memory tip: Think "Tumbling = Time-boxed + reTry + backTrack" - the four T's that distinguish it from a plain schedule trigger. If an exam question mentions backfill, concurrent execution limits, or retry within the same window, tumbling window is almost always the answer.

Topics

#Azure Data Factory#Tumbling Window Trigger#Pipeline Orchestration#Backfilling

Community Discussion

No community discussion yet for this question.

Full DP-203 Practice