SnowPro DEA-C02 Sample Questions with Explanations

These original sample questions are designed to help you check how the exam topics appear in decision-style prompts. They are not taken from the live exam.

Use these sample questions as a guided self-assessment for SnowPro Advanced: Data Engineer (DEA-C02) topics such as staged ingestion, Snowpipe Streaming, streams, tasks, dynamic tables, sharing, replication, warehouse fit, and performance evidence. The prompts emphasize pipeline responsibility and Snowflake-native design choices.

Where these questions fit in the DEA-C02 guide

The sample set below is part of the SnowPro DEA-C02 guide path:

• SnowPro certification hub
• DEA-C02 exam guide
• DEA-C02 cheat sheet
• DEA-C02 study plan
• DEA-C02 FAQ
• DEA-C02 resources
• DEA-C02 glossary

DEA-C02 Snowflake data engineering sample questions

Work through each prompt before opening the explanation. DEA-C02 questions usually reward answers that distinguish ingestion, transformation, orchestration, delivery, and performance diagnosis instead of choosing a feature from the wrong layer.

---

Question 1

Topic: Lower-latency ingestion choice

A telemetry producer needs to send event rows into Snowflake with lower latency than a staged-file polling pattern. The team wants an ingestion path designed for streaming writes rather than periodic file batches. Which option best fits?

• A. A monthly manual upload to an internal stage followed by COPY INTO.
• B. Snowpipe Streaming, when the producer and connector pattern fit streaming-row ingestion requirements.
• C. A zero-copy clone of the target table before every load.
• D. A secure share from the target table back to the source application.

Best answer: B

Explanation: The key clue is lower-latency streaming ingestion rather than staged-file batches. Snowpipe Streaming is designed for that row-oriented ingestion pattern when the source integration fits.

Why the other choices are weaker:

• A is a batch pattern and does not meet the latency requirement.
• C duplicates metadata for an existing object and does not ingest new events.
• D is a delivery/sharing pattern, not source ingestion.

What this tests: Snowpipe Streaming, staged loads, latency targets, and ingestion-pattern selection.

Related topics: Snowpipe Streaming; Ingestion; Latency; COPY INTO

---

Question 2

Topic: Streams, tasks, and dynamic tables

A pipeline must process only changed rows from a source table and run transformation logic on a schedule. Which Snowflake object pairing most directly matches change capture plus scheduled execution?

• A. A network policy to track changes and a role to schedule SQL.
• B. A warehouse clone to capture changes and a stage to execute SQL.
• C. A stream to track table changes and a task to run the processing SQL on a schedule.
• D. A secure share to capture changes and Time Travel to schedule SQL.

Best answer: C

Explanation: Streams expose change data for a table, while tasks schedule SQL execution. DEA-C02 frequently tests whether you identify which Snowflake object owns change tracking versus orchestration.

Why the other choices are weaker:

• A combines access-control concepts, not pipeline objects.
• B misuses cloning, stages, and warehouses.
• D confuses sharing and recovery with change processing and scheduling.

What this tests: streams, tasks, change capture, scheduling, and near real-time pipeline design.

Related topics: Streams; Tasks; Change data; Orchestration

---

Question 3

Topic: Sharing without copying data

A data provider wants a partner to query curated tables from the partner’s Snowflake account. The provider does not want to export files each night or create duplicate partner-owned storage copies. Which delivery model is strongest?

• A. A nightly COPY INTO unload to a partner stage because all cross-account delivery requires file copies.
• B. Sharing the provider’s ACCOUNTADMIN username and password with the partner.
• C. Creating a larger warehouse in the provider account so the partner can log into it directly.
• D. Secure Data Sharing or a Snowflake Marketplace-style sharing pattern, depending on distribution needs.

Best answer: D

Explanation: Secure sharing lets a provider expose governed live data to consumers without nightly file exports or duplicated storage copies. The clue is cross-account query access without copying.

Why the other choices are weaker:

• A may be an export pattern, but it contradicts the no-copy requirement.
• B is insecure and destroys accountability.
• C confuses compute sizing with governed data delivery.

What this tests: Secure Data Sharing, cross-account delivery, provider/consumer roles, and copy-versus-share decisions.

Related topics: Secure sharing; Provider; Consumer; Delivery

---

Question 4

Topic: Compute sizing with evidence

A transformation pipeline has become more expensive. Some tasks queue during busy windows, while other steps scan far more data than expected. What is the strongest first response?

• A. Use history, Query Profile, task run evidence, pruning behavior, and queueing signals to decide whether to tune SQL, adjust clustering or layout, change warehouse size, or separate workloads.
• B. Increase every warehouse to the largest size permanently.
• C. Disable monitoring because cost usually improves after the next billing cycle.
• D. Replace all SQL transformations with manual spreadsheet exports.

Best answer: A

Explanation: DEA-C02 performance answers should start from evidence. The scenario includes both queueing and scan-volume clues, so the fix may involve concurrency, workload isolation, SQL tuning, pruning, layout, or warehouse sizing.

Why the other choices are weaker:

• B may increase cost without fixing the real bottleneck.
• C removes the data needed to diagnose the problem.
• D abandons the governed Snowflake pipeline pattern.

What this tests: warehouse fit, Query Profile, task history, queueing, pruning, workload isolation, and cost control.

Related topics: Performance; Query Profile; Warehouses; Cost

Independent study note

Tech Exam Lexicon and IT Mastery are independent study tools. They are not affiliated with, endorsed by, or sponsored by Snowflake or any certification body.