Data Engineering ETL and ELT Pipelines 2

Question 1

In an ELT pipeline, where does the bulk of data transformation logic execute?

Accepted Answer

In the target data warehouse after loading

Answer

In a separate staging server before loading

Answer

On the source system before extraction

Answer

In the network transfer layer

Question 2

A pipeline reprocesses the entire source table every night instead of only changed rows. Which technique would reduce this load?

Accepted Answer

Incremental loading with change data capture

Answer

Full truncate-and-reload

Answer

Adding more network bandwidth

Answer

Disabling indexes

Question 3

Which scenario most favors ELT over traditional ETL?

Accepted Answer

A cloud warehouse with elastic, scalable compute like Snowflake or BigQuery

Answer

A legacy on-prem server with fixed, limited CPU

Answer

A real-time embedded device

Answer

A system with no storage for raw data

Question 4

What is the primary purpose of a staging area in an ETL pipeline?

Accepted Answer

A temporary landing zone for raw data before transformation

Answer

The final reporting layer for business users

Answer

A backup of production credentials

Answer

A cache for dashboard queries

Question 5

An idempotent pipeline run guarantees what behavior?

Accepted Answer

Running it multiple times produces the same final result

Answer

It runs faster on each subsequent execution

Answer

It never reads from the source twice

Answer

It always doubles the output rows

Question 6

Which approach best handles late-arriving data in a batch pipeline?

Accepted Answer

Reprocessing a lookback window of recent partitions

Answer

Ignoring any record past its scheduled window

Answer

Restarting the entire historical load nightly

Answer

Dropping the timestamp column

Question 7

What does schema drift refer to in a data pipeline?

Accepted Answer

Unexpected changes to source column names, types, or structure over time

Answer

Data physically moving between disk partitions

Answer

Gradual clock skew between servers

Answer

Slow degradation of network throughput