Data Engineering Orchestrating Data Workflows 2

Question 1

In Apache Airflow, what does a DAG represent?

Accepted Answer

A directed acyclic graph defining task dependencies

Answer

A database connection pool

Answer

A distributed file storage layer

Answer

A message queue partition

Question 2

Which Airflow concept controls how many task instances can run concurrently for a single DAG?

Accepted Answer

concurrency (max_active_tasks)

Answer

max_active_runs

Answer

retries

Answer

schedule_interval

Question 3

What is the purpose of an idempotent task in a data pipeline?

Accepted Answer

Re-running it produces the same result without side effects

Answer

It runs faster on each retry

Answer

It never needs to be scheduled

Answer

It bypasses dependency checks

Question 4

In a workflow scheduler, what is 'backfilling'?

Accepted Answer

Running a DAG for past dates that were missed or newly added

Answer

Deleting old task logs

Answer

Increasing worker memory

Answer

Compressing intermediate data

Question 5

Which statement best describes the difference between scheduling and orchestration?

Accepted Answer

Scheduling triggers jobs by time; orchestration manages dependencies and coordination

Answer

They are identical terms

Answer

Orchestration only runs single tasks

Answer

Scheduling handles error recovery exclusively

Question 6

In Airflow, what does an XCom enable?

Accepted Answer

Passing small amounts of data between tasks

Answer

Cross-DAG authentication

Answer

External cloud storage mounting

Answer

Compression of task logs

Question 7

What is a common reason to use a sensor in an orchestration tool?

Accepted Answer

To wait for an external condition like a file or partition to appear

Answer

To compress output files

Answer

To allocate GPU resources

Answer

To encrypt connection strings