GCP-PDE Google Cloud Professional Data Engineer
- Vendor
- Google Cloud
- Level
- Professional
- Questions
- 45
- Time
- 120 min
- Domains
- 5
- In the bank
- 285
Practice 285 questionsTake a timed mock examDay-by-day study plan
Guides for this exam
- 9 Question Patterns People Get Wrong Across the Google Cloud Exams, and the Rule Behind Each · 한국어
- GCP-PDE Study Guide: Five Phrases That Decide Most Data Engineer Questions, and BigQuery Under All of Them · 한국어
- Cloud SQL vs AlloyDB vs Spanner vs Bigtable vs Firestore vs BigQuery: How to Tell Which One a Question Wants · 한국어
The exam at a glance
| Level | Professional |
|---|---|
| Questions | 45 |
| Time | 120 minutes (about 2.7 minutes per question) |
| Passing mark | Not published by the vendor |
Domains and weights
| Domain | Weight | Practice questions here |
|---|---|---|
| D1 · Designing data processing systems | 22% | 63 |
| D2 · Ingesting and processing the data | 25% | 71 |
| D3 · Storing the data | 20% | 57 |
| D4 · Preparing and using data for analysis | 15% | 43 |
| D5 · Maintaining and automating data workloads | 18% | 51 |
55 of the 285 practice questions (19%) ask for more than one answer.
D1 · Designing data processing systems 22%
Design questions describe a data platform, often across several projects, and ask how to secure, govern or lay it out before any pipeline runs. Service accounts, IAM scope, encryption keys and data protection usually decide. Strong options give each job its own identity with only the access it needs; weaker ones reuse broad default accounts or copy data to work around a permission.
What to watch for
- Give each pipeline a dedicated service account and grant it roles on the exact datasets and buckets it touches, not on the whole project.
- Keep raw, curated and experimental data in separate datasets or projects so access boundaries can be proven to auditors.
- Customer-managed keys and Sensitive Data Protection answer compliance requirements; check which one the scenario actually asks for.
Tasks in the official exam guide (4)
D2 · Ingesting and processing the data 25%
Ingestion questions start from a source such as Pub/Sub, files or a database and a latency target, then ask for the simplest pipeline that meets it. Whether the data needs transformation often decides. Strong options use a direct subscription or managed connector when nothing needs changing, and Dataflow when it does; weaker ones add processing steps the requirement never mentions.
What to watch for
- When messages need no filtering or enrichment, a BigQuery subscription on the Pub/Sub topic is the pipeline with the fewest moving parts.
- Pull subscriptions suit private networks that cannot expose a public push endpoint.
- For troubleshooting, separate delivery problems (acknowledgement, backlog) from processing problems (windowing, late data) before choosing a fix.
Tasks in the official exam guide (3)
D3 · Storing the data 20%
Storage questions describe access patterns, retention and cost limits and ask where and how to keep the data. How often data is read again, and by whom, usually decides. Strong options match the storage class, table design or database to the access pattern; weaker ones pick the most powerful store or move data between services without need.
What to watch for
- Autoclass manages Cloud Storage classes for unpredictable access; know which bucket settings stop it from being enabled.
- In BigQuery, partitioning and clustering on the columns that filters use cut both cost and query time.
- Pick Bigtable, Spanner or Cloud SQL by access pattern and consistency needs, not by data size alone.
Tasks in the official exam guide (4)
D4 · Preparing and using data for analysis 15%
Analysis questions ask how to prepare data so that analysts, dashboards or models get correct results, increasingly with embeddings and vector search in BigQuery. Accuracy requirements and how often a query runs usually decide. Strong options choose exact or approximate methods on purpose; weaker ones trade correctness for speed the scenario never asked for.
What to watch for
- When a missed result is a real risk and the table is small, an exact search beats an approximate vector index.
- Embeddings from different models are not comparable: re-embed the stored data when the model changes.
- Shared, governed views and authorized datasets let analysts query without copies of sensitive tables.
Tasks in the official exam guide (3)
D5 · Maintaining and automating data workloads 18%
Operations questions describe a running workload that costs too much, fails or needs scheduling, and ask for the smallest change that fixes it. Idle resources, retries and monitoring gaps usually decide. Strong options automate the control with a managed feature such as scheduled deletion or an orchestrator; weaker ones rewrite the job or depend on people remembering to act.
What to watch for
- Clusters that sit idle most of the day point to ephemeral or serverless Dataproc, or to scheduled deletion.
- Cloud Composer or Workflows orchestrate dependencies; a cron trigger alone cannot handle retries and ordering.
- Alert on the symptom users feel, such as data freshness or backlog, not only on machine metrics.
Tasks in the official exam guide (5)
A study order that works
The 22-day study plan for GCP-PDE takes the domains in this order. Each day has a short reading, key terms and a quiz from the bank.
- Days 1–4: D1 · Designing data processing systems
- Days 5–10: D2 · Ingesting and processing the data
- Days 11–14: D3 · Storing the data
- Days 15–17: D4 · Preparing and using data for analysis
- Days 18–21: D5 · Maintaining and automating data workloads
- Day 22: Final Review: Decision Cheat-Sheet Across the Five Domains and Test Day
Recent changes to these questions
- : 4 answer choice sets changed, 4 answer keys corrected, 4 explanations rewritten
- : 60 questions added
- : 2 answer choice sets changed, 2 answer keys corrected, 2 explanations rewritten
- : Exam added to the bank (225 questions added)
About these questions
Every question is original, written from the public exam guide and checked against the vendor's own documentation. None comes from a real exam. Each one has an explanation of why the answer is right and why the other options are wrong. How the questions are written.