Examkin

Question

Difficulty: HardPlanning Database Options and Persistent Disk Types

A financial services company is building a high-volume transaction system on Google Cloud. The core workload requires a fully managed, globally scalable relational database that guarantees multi-region ACID compliance and supports standard SQL queries. Simultaneously, a data-processing microservice running on Compute Engine needs high-throughput temporary block storage for ephemeral scratch data, requiring maximum IOPS where data survival across instance termination is unnecessary. Which combination of Google Cloud database and block storage solutions should be selected?

  1. Cloud Spanner for the core transaction database, and Local SSD for the ephemeral scratch storageAnswer
  2. B
    Cloud SQL with cross-region read replicas for the core transaction database, and Balanced Persistent Disk (pd-balanced) for the ephemeral scratch storage
  3. C
    Cloud Bigtable for the core transaction database, and SSD Persistent Disk (pd-ssd) for the ephemeral scratch storage
  4. D
    Cloud Spanner for the core transaction database, and Standard Persistent Disk (pd-standard) for the ephemeral scratch storage

Answer

Cloud Spanner for the core transaction database, and Local SSD for the ephemeral scratch storage
The requirement calls for a globally scalable relational database with multi-region ACID compliance and standard SQL support, which specifically maps to Cloud Spanner. For the Compute Engine microservice requiring maximum IOPS ephemeral block storage where data does not need to persist after VM termination, Local SSD is the recommended solution as it is physically attached to the server host.

Step-by-Step Solution

1
Analyze database requirements for the core transaction workload.
Identified requirements for a fully managed relational database with standard SQL support, multi-region ACID transactions, and horizontal scaling.
Cloud Spanner is Google Cloud's only fully managed relational database that provides global scale, multi-region ACID compliance, and horizontal write scalability.
2
Analyze block storage requirements for the Compute Engine microservice.
Identified requirements for maximum IOPS, microsecond/sub-millisecond latency, ephemeral scratch space, and no need for persistence across instance termination.
Local SSDs are physically attached to the host running the VM instance, offering high IOPS and ultra-low latency ideal for temporary scratch disk space.
3
Evaluate the architectural combination against Google Cloud best practices.
Combining Cloud Spanner with Local SSD satisfies both requirements with minimal operational overhead.
This combination directly fulfills all throughput, consistency, scale, and performance criteria without over-provisioning network storage.

Key Concept

Selecting managed database and persistent storage options based on transaction consistency, geographic scope, and IOPS requirements
Rate this question

Topics

Planning Database Options and Persistent Disk Types

More questions from this topic