Google Cloud Professional Cloud Database Engineer
225 practice questions
Last reviewed: April 2026
Personal notes and resource links for your study journey
Filter by Certification
The Google Cloud Professional Cloud Database Engineer (PCDE) validates the ability to design, manage, migrate, and operate database solutions on Google Cloud — covering Cloud SQL (MySQL, PostgreSQL, SQL Server), AlloyDB for PostgreSQL, Spanner, Bigtable, Firestore, Memorystore (Redis and Memcached), and BigQuery as an analytical store. The exam emphasizes high availability, disaster recovery, performance tuning, replication topologies, migration patterns (Database Migration Service, Datastream), and the operational realities of running OLTP and analytical workloads at scale. PCDE is the GCP analog of AWS Database Specialty and Azure DP-300 / DP-420 combined. It targets database engineers, DBAs moving to cloud, and senior platform engineers responsible for production data layers.
Largest domain at 32%. HA topologies (regional, multi-regional), read replicas, failover, point-in-time recovery, choosing between Cloud SQL / AlloyDB / Spanner for OLTP, Bigtable for wide-column, Firestore for document, Memorystore for cache.
Polyglot persistence patterns, cross-database querying with federated queries, CDC with Datastream, integrating operational stores with BigQuery analytics, IAM and audit logging across database surfaces. 25%.
Database Migration Service for MySQL / PostgreSQL / SQL Server / Oracle homogeneous and heterogeneous migrations, schema conversion, downtime-minimization patterns, validation. 23%.
IaC with Terraform, configuration management, capacity planning, autoscaling for AlloyDB and Spanner, query insights, operational monitoring with Cloud Operations. 20%.
Services you'll encounter on the exam and why each one matters.
Fully managed relational database for PostgreSQL, MySQL, and SQL Server with automated backups, point-in-time recovery, read replicas, and Enterprise Plus tiers.
Why it's on the exam: Domain 1 (Design scalable/HA database solutions) anchors most relational-DB scenarios — sizing, replica topology, and zonal vs regional HA selection.
Globally distributed, strongly consistent relational database with horizontal scale, schema-aware secondary indexes, and 99.999% multi-region SLA.
Why it's on the exam: Domain 1 + Domain 4 (Deploy scalable/HA databases) frame Spanner as the canonical answer for write-heavy, globally consistent workloads.
Petabyte-scale wide-column NoSQL store with single-digit-millisecond reads, used for time-series, IoT, ad-tech, and large analytical fact tables.
Why it's on the exam: Domain 1 distinguishes Bigtable from Spanner/Firestore — high-throughput, low-latency, sparse-row workloads are the recurring Bigtable signal.
Serverless document database with real-time sync, Firestore-native and Datastore modes, and automatic multi-region replication.
Why it's on the exam: Domain 1 covers Firestore as the managed NoSQL document option for mobile/web back ends, contrasted against Cloud SQL and Spanner.
PostgreSQL-compatible managed database with disaggregated storage, columnar acceleration, ML integration, and up to 4× vanilla PostgreSQL transactional throughput.
Why it's on the exam: Domain 1 + Domain 4 contrast AlloyDB against Cloud SQL for PostgreSQL on HA SLA, scale ceiling, and analytical-on-OLTP hybrid scenarios.
Per-engine query performance tooling — Query Insights for Cloud SQL/AlloyDB, Query Stats and Lock Insights for Spanner — surfacing slow statements, lock waits, and plan regressions.
Why it's on the exam: Domain 2 (Manage multi-database solutions) tests how to diagnose query-plan regressions and contention across heterogeneous engines.
Managed service for homogeneous (PG→Cloud SQL) and heterogeneous (Oracle→PostgreSQL) migrations with continuous CDC replication and minimal-downtime cutover.
Why it's on the exam: Domain 3 (Migrate data solutions) is anchored on DMS — assessment, schema conversion, and CDC cutover patterns are recurring scenario questions.
Fully managed in-memory cache for Redis (with read replicas, RDB/AOF persistence) and Memcached, used as caching layer or session store in front of relational DBs.
Why it's on the exam: Domain 4 covers Memorystore as the canonical write-through/cache-aside answer for offloading read pressure from Cloud SQL and Spanner.
Serverless multi-cloud analytical data warehouse with separation of storage and compute, materialized views, BI Engine, and BigQuery ML for in-warehouse modeling.
Why it's on the exam: Domain 2 frames BigQuery as the OLAP destination when reporting/aggregation pressure should not hit transactional engines.
Serverless change-data-capture service streaming row-level changes from Oracle, MySQL, PostgreSQL, and SQL Server into BigQuery, Cloud Storage, or Cloud Spanner.
Why it's on the exam: Domain 3 cites Datastream for low-latency CDC into a target during migration and for keeping analytics tables fresh post-cutover.
Regional persistent disk synchronous HA for Cloud SQL with sub-minute failover, plus cross-region and cross-zone read replicas for read scale-out and DR.
Why it's on the exam: Domain 4 (Deploy scalable/HA databases) repeatedly tests RTO/RPO tradeoffs — sync HA, async cross-region replicas, and promotion procedures.
Multi-region instance configurations (e.g. nam-eur-asia1, nam6) with synchronous replication plus point-in-time recovery and managed backup/restore.
Why it's on the exam: Domain 1 + Domain 4 frame Spanner multi-region as the answer to "five-nines SLA + RPO=0" scenarios where regional Cloud SQL can't meet the bar.
In-memory columnar accelerator inside AlloyDB that auto-populates from access patterns and transparently rewrites analytical predicates against row store.
Why it's on the exam: Domain 2 cites AlloyDB columnar as the answer when analytical queries co-exist with OLTP and a full move to BigQuery isn't warranted.
BigQuery Omni runs BigQuery on AWS S3 and Azure Blob; federated queries reach into Cloud SQL, Spanner, Bigtable, and external tables without bulk ETL.
Why it's on the exam: Domain 2 (Manage multi-database solutions) tests cross-engine query patterns and when to federate vs. replicate for analytics.
Sidecar process that wraps the Cloud SQL connection with IAM-authenticated, TLS-encrypted tunneling — no public IP, no client certificate management.
Why it's on the exam: Domain 4 + Domain 2 cite the Auth Proxy as the canonical answer for least-privilege, network-locked-down Cloud SQL access from GKE/Cloud Run/serverless.
Managed pub/sub messaging used to fan database change events (via DMS, Datastream, or Spanner change streams) to downstream consumers and event-driven workloads.
Why it's on the exam: Domain 2 frames Pub/Sub as the change-event bus that decouples database writes from downstream caches, search indexes, and analytics pipelines.
Project- and instance-level Cloud IAM roles for database administration plus IAM database authentication that maps Google identities to PostgreSQL/MySQL users.
Why it's on the exam: Domain 4 (Deploy scalable/HA databases) tests least-privilege patterns — IAM DB auth removes long-lived passwords and ties access to Workforce Identity.
Customer-managed encryption keys layered on top of Google-managed encryption for Cloud SQL, Spanner, AlloyDB, Bigtable, Firestore, and backup storage.
Why it's on the exam: Domain 1 + Domain 4 expect CMEK for regulated workloads — key rotation, key import, and HSM-backed keys are recurring exam knobs.
Immutable audit trail of admin and data-access API calls across managed databases; Data Access logs capture row-level read/write activity (opt-in due to volume).
Why it's on the exam: Domain 4 compliance scenarios require Data Access logs to answer "who queried this PII row" and "when was this DB modified."
Service perimeter that blocks data exfiltration from managed databases (Cloud SQL, Spanner, BigQuery, Bigtable) to outside-perimeter projects/identities, even with valid IAM.
Why it's on the exam: Domain 4 tests perimeter design for regulated databases — VPC-SC complements IAM by preventing legitimate-credential-driven exfil paths.
$130k–$180k–$265k USD annual
Range reflects US-based senior database engineers and DBAs where GCP is the primary platform. FAANG L5 database engineer TC clears $300k; specialty Spanner / AlloyDB roles trend toward the high end. The cloud-DBA candidate pool on GCP is genuinely small, which gives PCDE holders strong negotiating leverage.
Source: levels.fyi 2025–2026 (Google L4–L5 database engineers, FAANG and GCP-shop unicorn senior DBA / DRE), U.S. BLS OEWS May 2024 (15-1245 database administrators and architects, 15-1244 network and computer systems administrators). Figures are approximate; actual compensation depends on role, region, and experience.
PCDE is the most niche of the GCP Professional credentials by absolute job-board volume, but holders consistently report strong recruiter response because the GCP database-specialist candidate pool is tiny. Demand concentrates at large enterprises migrating Oracle / SQL Server workloads to GCP, fintech and gaming companies running Spanner, ad-tech and IoT platforms running Bigtable, and Google Cloud partners with database-modernization practices. Google's push of AlloyDB as an enterprise PostgreSQL alternative through 2024–2026 created additional pull on this credential. PCDE pairs naturally with Professional Data Engineer (PDE) for a broader data-engineering profile.
There are no formal prerequisites. Google recommends three or more years of industry experience and one or more years designing and managing database solutions on Google Cloud. In practice, PCDE is not a credible first GCP cert — successful candidates have meaningful traditional DBA or database-engineering experience (Oracle, SQL Server, PostgreSQL, MySQL, or NoSQL).
The Associate Cloud Engineer (ACE) is a common stepping stone but a strong on-prem DBA background often substitutes well. Comfort with SQL performance tuning, replication topologies, backup / restore mechanics, and at least one of MySQL or PostgreSQL administration is effectively required. The official Cloud Database Engineer Learning Path on Google Cloud Skills Boost (around 40–60 hours) covers the curriculum; most successful candidates also build a Database Migration Service migration lab end-to-end.
PCDE is rated professional and is moderately hard — the breadth across Cloud SQL, AlloyDB, Spanner, Bigtable, Firestore, and Memorystore is the main challenge, more than depth in any one. Plan on 80–120 hours of study over 8–12 weeks if PCDE is your first GCP professional cert, or 40–70 hours over 4–6 weeks if you already hold ACE plus production DBA experience. The exam is 50–60 multiple-choice / multiple-select questions in 120 minutes, delivered through Pearson VUE (Google migrated from Kryterion / Webassessor in early 2026).
The most common stumbling block is choosing between Cloud SQL, AlloyDB, and Spanner for OLTP scenarios where multiple options are technically viable — Google's "preferred" answer often hinges on scale ceiling, regional vs. global writes, and cost rather than pure technical fit. The second stumbling block is Database Migration Service mechanics for heterogeneous migrations (Oracle to PostgreSQL especially). Google does not publish numeric scores — only pass/fail. The credential is valid for two years and recertification requires re-passing the current exam.
Current exam guide refreshed in mid-2024 to add AlloyDB Omni, expanded Spanner Granular Instance coverage, and updated Database Migration Service heterogeneous migration scenarios.
Initial general availability — the newest GCP Professional credential, introduced as Google expanded its managed-database portfolio with AlloyDB.
PCDE (Google Cloud Professional Cloud Database Engineer) is a a challenging, scenario-heavy exam that requires deep hands-on experience and the ability to make architectural trade-off decisions Professional-level exam. Most candidates need 150–300 hours of study spread over 3–6 months for professional and expert-level exams. These exams typically expect prior associate-level proficiency. Most candidates who score consistently above the passing threshold on practice exams pass on their first attempt.
Most candidates need 150–300 hours of study spread over 3–6 months for professional and expert-level exams. These exams typically expect prior associate-level proficiency. Time-to-pass varies widely by prior experience. Engineers with hands-on production experience in the underlying technology typically need less; candidates new to the platform should plan toward the upper end of that range.
PCDE is a recognized credential in the GCP ecosystem and signals validated knowledge to employers, recruiters, and clients. Whether it is worth the time and fee for you depends on your role and goals — it tends to pay off most for cloud engineers, architects, and consultants who work with GCP day-to-day or want to move into roles that do.
The passing score for PCDE is Not published. The exam contains 50 questions and lasts 2 hr.
The PCDE exam fee is $200 USD. Fees are set by GCP and may vary by region; always confirm the current price on the official GCP certification page before booking.
Google Cloud Professional certifications are valid for 2 years. Recertify by re-passing the current version of the exam.
Yes. You can take the exam online (proctored via the provider's secure browser, available 24/7 in most regions) or at an in-person Pearson VUE test center during business hours. Both formats use the same questions, time limit, and passing score.
CertLabPro provides 15 study modes across the practice question bank for PCDE. The exam-simulation mode mirrors the real exam: 50 questions in 2 hr, with the same passing threshold of Not published. Browse mode lets you read every Q&A statically.