Home › GCP Data Engineer › cloudcomputing › gcpdeprocess › Which Dataflow feature should a Data Engineer en…
Which Dataflow feature should a Data Engineer enable to BACKFILL or REPROCESS late-arriving data from Pub/Sub WITHOUT re-emitting already-published Pub/Sub messages?
ADataflow streaming jobs with snapshots (and Pub/Sub seek for replay where needed)
BDisabling exactly-once processing
CCloud Memorystore
DCloud DNS
Answer & Solution
Correct answer: A. Dataflow streaming jobs with snapshots (and Pub/Sub seek for replay where needed)
Dataflow snapshots + Pub/Sub seek enable reprocessing without re-publishing (per PDE §2.2). The other options aren't backfill features.
Related questions
Which Dataflow feature lets a Data Engineer reuse a published pipeline as a parameterised,Which Dataproc feature should a Data Engineer use to run a Spark job ONLY when work existsWhich GCP service lets a Data Engineer write streaming SQL queries directly against a Pub/Which Pub/Sub delivery model should a Data Engineer use when subscribers should pull messaWhich BigQuery feature lets a Data Engineer run ANALYTICAL QUERIES across data still livinWhich GCP service should a Data Engineer use for lightweight, serverless workflow orchestrWhich GCP service should a Data Engineer use to orchestrate Cloud Dataflow / BigQuery / DaWhich Apache Beam concept handles events arriving AFTER their window closes — letting the