You need ads data to serve AI models and historical data for analytics. Longtail and outlier data points need to be identified. You want to cleanse the data in near- real time before running it throug

Question

Options

• AUse Cloud Storage as a data warehouse, shell scripts for processing, and BigQuery to create views for desired datasets.
• BUse Dataflow to identify longtail and outlier data points programmatically, with BigQuery as a sink.
• CUse BigQuery to ingest, prepare, and then analyze the data, and then run queries to create views.
• DUse Cloud Composer to identify longtail and outlier data points, and then output a usable dataset to BigQuery.

• A
  14% (3)
• B
  73% (16)
• C
  9% (2)
• D
  5% (1)

Explanation

B is correct because Dataflow (built on Apache Beam) is purpose-built for near-real-time stream processing, making it ideal for programmatically identifying longtail and outlier data points during ingestion before the data reaches AI models - with BigQuery as the sink covering both the AI serving and historical analytics requirements.

A is wrong because Cloud Storage is an object store, not a data warehouse, and shell scripts are neither scalable nor capable of near-real-time processing at the level this scenario demands.

C is wrong because BigQuery is an OLAP data warehouse optimized for SQL-based batch analytics - it lacks the streaming pipeline capabilities needed to cleanse data before it reaches AI models in near-real-time.

D is wrong because Cloud Composer (managed Apache Airflow) is a workflow orchestration tool for scheduling and coordinating pipelines, not a data processing engine - it cannot itself identify outliers or cleanse streaming data.

Memory tip: When you see "near-real-time cleansing" + "programmatic transformations" on the exam, think Dataflow - it's the GCP streaming/batch processing engine. Cloud Composer schedules work; Dataflow does the work.

No community discussion yet for this question.