A global pharmaceutical enterprise is architecting a mission-critical clinical trial telemetry platform on Google Cloud. The system requires a relational database that guarantees automatic zonal failover with minimal Recovery Time Objective (RTO) within a single region. Additionally, the on-premises laboratory network requires a high-availability, dedicated 10 Gbps private connection to the Google Virtual Private Cloud (VPC) with high throughput and low latency. Which TWO architectural components should the cloud architect select to meet these technical and availability requirements? (Select TWO.)
- Provision Cloud SQL configured with regional High Availability (HA) using a primary instance in one zone and a standby instance in a secondary zone within the target region.Answer
- Establish Dedicated Interconnect connections with redundant circuits across two edge locations to connect the on-premises network to Google Cloud.Answer
- CDeploy a multi-region Cloud Spanner instance as the relational database engine for the single-region clinical trial platform.
- DConfigure standard HA VPN tunnels over the public internet to deliver 10 Gbps hybrid network throughput.
- EConfigure backend application compute autoscaling strictly on CPU utilization metrics to handle incoming network-heavy laboratory payloads.
Answer
The architect should deploy Cloud SQL in a regional High Availability configuration and establish Dedicated Interconnect redundant connections for 10 Gbps dedicated hybrid networking.
To achieve single-region database high availability with automatic zonal failover, Cloud SQL HA (High Availability) mode provisions a primary and synchronous standby instance across distinct zones. For hybrid network connectivity needing 10 Gbps dedicated bandwidth with enterprise SLA, Dedicated Interconnect is required because Cloud VPN cannot deliver 10 Gbps per link.
Step-by-Step Solution
Key Concept
Designing high-availability cloud infrastructure requires selecting compute, database, and network services that match specific RTO, SLA, and throughput constraints without over-engineering.