Examkin

Question

Difficulty: MediumDesigning Infrastructure for Technical Requirements and High Availability

A financial technology company is architecting a core payment fraud detection service on Google Cloud. The workload processes stateless HTTP requests originating from client applications globally and requires a service availability SLA of 99.99%. The underlying relational transaction database must support high-throughput global writes, maintain synchronous multi-region consistency, and guarantee zero Recovery Point Objective (RPO) and immediate Recovery Time Objective (RTO) in the event of an entire Google Cloud region failure. Which infrastructure design meets these technical and high availability requirements while minimizing operational management overhead?

  1. Deploy the stateless service on Cloud Run across multiple regions behind an External HTTP(S) Load Balancer, and use a multi-region Cloud Spanner database instance.Answer
  2. B
    Deploy the stateless service on a regional Google Kubernetes Engine (GKE) Autopilot cluster, and use a multi-region Cloud Spanner database instance.
  3. C
    Deploy the stateless service on Cloud Run across multiple regions behind an External HTTP(S) Load Balancer, and use Cloud SQL for PostgreSQL with cross-region read replicas.
  4. D
    Deploy the stateless service on Cloud Run across multiple regions, and use a single-region Cloud SQL instance with High Availability (HA) enabled across two zones.

Answer

Deploying the stateless application on multi-region Cloud Run integrated with a multi-region Cloud Spanner database fulfills all high availability, multi-region failover, zero RPO, and minimal management overhead criteria.
Combining multi-region Cloud Run with a multi-region Cloud Spanner instance delivers a fully managed architecture capable of surviving a complete regional failure without data loss (RPO = 0) or manual intervention. Cloud Run minimizes management overhead for stateless web services, while Cloud Spanner provides multi-region synchronous replication for transactional relational data.

Step-by-Step Solution

1
Evaluate compute requirements for stateless HTTP processing and operational overhead.
Cloud Run provides serverless container deployment that automatically scales to zero and scales out across multiple regions without requiring cluster administration, making it preferred over GKE for simple stateless microservices.
Minimizing operational complexity requires avoiding container cluster management when a serverless option satisfies all scaling and availability targets.
2
Evaluate relational database requirements for high availability and disaster recovery.
Cloud Spanner configured with a multi-region instance deployment provides synchronous replication across regions with Paxos consensus, ensuring strict serializability, zero RPO, and automatic multi-region failover.
Cloud SQL relies on asynchronous replication across regions, which risks data loss (violating RPO = 0) during a regional disaster.

Key Concept

Multi-region high availability architectures combining serverless compute (Cloud Run) and globally synchronous relational storage (Cloud Spanner)
Estimated Time:1m 30s
Rate this question

Topics

Designing Infrastructure for Technical Requirements and High Availability