You work with a data engineering team that has developed a pipeline to clean your dataset and save it in a Cloud Storage bucket. You have created an ML model and want to use the data to refresh your m

Question

Options

• AConfigure your pipeline with Dataflow, which saves the files in Cloud Storage.
• BUse App Engine to create a lightweight python client that continuously polls Cloud Storage for
• CConfigure a Cloud Storage trigger to send a message to a Pub/Sub topic when a new file is
• DUse Cloud Scheduler to schedule jobs at a regular interval.

• A
  16% (3)
• B
  5% (1)
• C
  68% (13)
• D
  11% (2)

Explanation

The requirement is event-driven: trigger a Kubeflow Pipelines job automatically when a new file appears in Cloud Storage. Option C is correct because Cloud Storage can publish object-creation notifications directly to a Pub/Sub topic; a subscriber (such as a Cloud Function) then receives the message and programmatically submits a Kubeflow Pipelines run on GKE. This is fully event-driven, serverless, and requires no polling. Option A misframes the problem-Dataflow is for data processing, not for triggering training jobs, and does not explain the trigger mechanism. Option B (polling with App Engine) is inefficient, wasteful, and not a CI/CD best practice. Option D (Cloud Scheduler) is time-based and would trigger jobs on a fixed schedule regardless of whether new data exists, which does not satisfy the 'as soon as new data is available' requirement.

No community discussion yet for this question.