Examkin

Question

Difficulty: MediumDesigning Infrastructure for Technical Requirements and High Availability

An international hospitality enterprise is architecting its real-time global reservation system on Google Cloud. The system must maintain an availability target of 99.99%, a Recovery Time Objective (RTO) of under 1 minute, and a Recovery Point Objective (RPO) of zero for transactional booking data across North America and Europe. The application backend consists of lightweight, stateless containerized HTTP microservices subject to unpredictable traffic bursts. Which infrastructure design meets these high-availability and technical requirements while minimizing operational overhead?

  1. Deploy the stateless microservices on Cloud Run across multiple regions behind a Global External Application Load Balancer, and store transactional reservation data in a Cloud Spanner multi-region instance configuration.Answer
  2. B
    Deploy regional Google Kubernetes Engine (GKE) clusters in both continents managed by Anthos Service Mesh, and store reservation data in a Cloud Spanner multi-region instance configuration.
  3. C
    Deploy the stateless microservices on Cloud Run across multiple regions behind a Global External Application Load Balancer, and store reservation data in a primary Cloud SQL for PostgreSQL instance with a cross-region read replica.
  4. D
    Deploy the stateless microservices on Compute Engine regional Managed Instance Groups (MIGs), backing up database transactions via hourly Cloud Storage exports to be restored in a secondary region upon failover.

Answer

Deploying the stateless microservices on Cloud Run across multiple regions behind a Global External Application Load Balancer combined with a Cloud Spanner multi-region instance configuration best meets all availability, RTO, RPO, and operational requirements.
The combination of Cloud Run and Cloud Spanner multi-region fulfills all technical constraints. Cloud Run handles stateless containerized HTTP microservices with automatic regional scaling and zero management of Kubernetes clusters. Cloud Spanner multi-region uses synchronous multi-region replication via Paxos to deliver 99.999% availability, zero RPO (no transaction loss), and automatic sub-minute failover (RTO < 1 min).

Step-by-Step Solution

1
Evaluate compute options for stateless HTTP microservices with variable traffic and low operational overhead.
Cloud Run provides serverless container hosting that automatically scales from zero to meet traffic spikes without managing node pools or cluster control planes.
Choosing serverless over GKE avoids infrastructure management overhead for simple stateless services.
2
Evaluate database options against high availability (99.99%), zero RPO, and sub-minute RTO multi-region requirements.
Cloud Spanner multi-region configurations utilize synchronous Paxos consensus across regional zones to deliver multi-region active-active read/write capabilities with zero data loss (RPO = 0) and seamless failover.
Cloud SQL relies on asynchronous cross-region replication which incurs data loss lag (RPO > 0), whereas Cloud Spanner natively provides synchronous multi-region consistency.

Key Concept

Designing multi-region active-active architectures combining serverless compute and globally consistent synchronous databases for high availability and strict RTO/RPO targets.
Rate this question

Topics

Designing Infrastructure for Technical Requirements and High Availability