Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Professional Cloud Database Engineer practice bank has an additional 769 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
A financial services company is migrating a mission-critical transactional database to Google Cloud and wants to minimize operational overhead. The database team has the following requirements:
Which database solution should the company implement?
Spanner is a fully managed, globally distributed relational database service built to provide strong transactional consistency, horizontal scalability, and high availability. In a multi-region instance configuration, Google Cloud automatically distributes your database across multiple geographic regions—typically maintaining two read-write regions and one witness region containing a witness replica for quorum voting.
Unlike unmanaged Compute Engine deployments that demand complex cluster orchestration or regional managed databases that rely on asynchronous cross-region read replicas, Spanner multi-region natively delivers active-active writes, automated failover, and minimal operational toil in a single managed service.
In an unmanaged Compute Engine deployment, you can offload data replication from the database engine to Google Cloud's block storage layer. Regional Persistent Disks (or Hyperdisk Balanced High Availability) synchronously replicate disk blocks across two zones in a region, while Persistent Disk Asynchronous Replication (PD Async Replication) asynchronously replicates block storage across regions using consistency groups.
The organization was migrating legacy self-managed database workloads to VMs and wanted infrastructure-level active-passive storage replication without relying on database-level replication software.
Cloud SQL is a fully managed relational database service for MySQL, PostgreSQL, and SQL Server. Enabling regional high availability (HA) deploys a primary instance in one zone and a synchronously replicated standby instance in another zone within the same region. Adding a cross-region read replica creates an asynchronously updated copy in a separate geographic region for read scaling and disaster recovery (DR).
The company required a managed regional relational database with automatic zonal failover, could tolerate an active-passive DR model with some replication lag across regions, and wanted to avoid managing VMs.
Deploying self-managed database instances (such as SQL Server Always On availability groups, MySQL, or PostgreSQL) on Compute Engine virtual machines (VMs) gives administrators full control over the operating system and database engine. High availability and disaster recovery are achieved by manually configuring application-level replication between primary, secondary, and cross-region nodes alongside cluster management tools like Pacemaker or Windows Server Failover Clustering (WSFC).
The organization needed custom database engine extensions, OS-level access, or specific third-party licensing configurations that cannot run on a managed database service.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated Validated