Examkin

Soru

Zorluk: OrtaSelecting and Designing Storage and Database Resources

A financial technology startup is designing the transactional backend for a peer-to-peer payment processing service operating exclusively within a single Google Cloud region. The service requires a relational database that enforces strict ACID compliance and complex foreign key relationships. The infrastructure must provide high availability with automated failover and zero data loss (RPO = 0) during a single zonal outage. Peak throughput is estimated at 2,500 write transactions per second. Global multi-region scaling is explicitly out of scope, and minimizing operational cost is a primary constraint. Which database solution should the cloud architect choose?

  1. Provision a Cloud SQL for PostgreSQL instance with High Availability (HA) enabled across two zones.Cevap
  2. B
    Provision a regional Cloud Spanner instance with multi-zone node deployment.
  3. C
    Provision a Cloud Bigtable cluster with multi-cluster replication across two zones in the region.
  4. D
    Provision a Cloud Storage Dual-Region bucket formatted with CSEK encryption for direct transaction persistence.

Cevap

Provision a Cloud SQL for PostgreSQL instance with High Availability (HA) enabled across two zones.
Cloud SQL for PostgreSQL with High Availability (HA) enabled fulfills all requirements: full ACID compliance, relational foreign key support, automatic failover across zones within a single region with zero data loss (RPO = 0), and a cost-effective price point suitable for single-region deployment.

Adım Adım Çözüm

1
Analyze data model and transactional constraints.
Relational data model with strict ACID compliance and foreign key constraints requires a relational RDBMS (Cloud SQL or Cloud Spanner). NoSQL options like Cloud Bigtable or object storage like Cloud Storage are eliminated.
Live financial transaction workloads requiring complex queries and relational constraints must use a relational database.
2
Evaluate geographic scale and cost constraints.
Workload is strictly single-region, and cost minimization is required.
Cloud Spanner is designed for massive global horizontal scale and multi-region synchronization. For single-region workloads with standard throughput, Cloud SQL provides the required capabilities at significantly lower cost.
3
Verify High Availability configuration.
Enabling High Availability (HA) on Cloud SQL configures synchronous regional persistent disk replication to a standby instance in a second zone, satisfying the RPO = 0 requirement.
Cloud SQL HA ensures automatic failover without data loss during a zonal outage.

Anahtar Kavram

Selecting Cloud SQL over Cloud Spanner for single-region relational database workloads to satisfy ACID requirements while optimizing cost.
Bu soruyu puanla

Konular

Selecting and Designing Storage and Database Resources