CLOUD-NETWORK practice question 14 of 30

You need to create the technical architecture for hybrid connectivity from your data center to Google Cloud This will be managed by a partner. You want to follow Google-recommended practices for production-level applications. What should you do?

Full set: 233 questions
Answer options
A: Ask the partner to install two security appliances in the data center. Configure one VPN connection from each of these devices to Google Cloud, and ensure that the VPN devices on-premises are in separate racks on separate power and cooling systems.
B: Configure two Partner Interconnect connections in one metropolitan area (metro). Make sure the Interconnect connections are placed in different metro edge availability domains. Configure two VLAN attachments in a single region, and configure regional dynamic routing on the VPC
C: Configure two Partner Interconnect connections in one metro and two connections in another metro Make sure the Interconnect connections are placed in different metro edge availability domains. Configure two VLAN attachments in one region and two VLAN attachments in another region, and configure global dynamic routing on the VPC Correct answer
D: Configure two Partner Interconnect connections in one metro and two connections in another metro. Make sure the Interconnect connections are placed in different metro edge availability domains. Configure two VLAN attachments in one region and two VLAN attachments in another region, and configure regional dynamic routing on the VPC.
Correct Answer:
C. Configure two Partner Interconnect connections in one metro and two connections in another metro Make sure the Interconnect connections are placed in different metro edge availability domains. Configure two VLAN attachments in one region and two VLAN attachments in another region, and configure global dynamic routing on the VPC
Explanation

For production-level applications requiring high availability, Google's recommended practice is to establish full redundancy across both geographic locations and Google Cloud regions. This involves configuring two Partner Interconnect connections in one metropolitan area (metro) and two more in a second, geographically separate metro. The connections in each metro must be in different edge availability domains (EADs). Correspondingly, VLAN attachments should be created in two different Google Cloud regions. Using global dynamic routing on the VPC is critical, as it allows Cloud Routers in one region to learn routes from the other, enabling automatic failover if an entire metro or region becomes unavailable.

Why Incorrect

A. A VPN-based solution is generally not the recommended practice for high-throughput, production-level, partner-managed connectivity compared to Partner Interconnect.
B. This topology provides redundancy within a single metro and region but lacks protection against a larger-scale metro or regional failure.
D. Using regional dynamic routing prevents the VPC from learning routes from the other region, which negates the benefit of the multi-region setup for failover.

References

1. Google Cloud Documentation, "Partner Interconnect topologies". Section: "Topology for production-level applications". This section explicitly describes the recommended architecture: "four VLAN attachments—two per region... four Partner Interconnect connections—two per metropolitan area (metro)... The connections are in different edge availability domains... you enable global routing in the VPC network."

2. Google Cloud Documentation, "VPC network dynamic routing modes". This document explains that "With global routing mode, Cloud Routers can advertise subnets from their local region... and also advertise subnets from other regions." This is essential for the multi-region failover described in option C.

Scroll to Top