GCP Professional Cloud Architect Practice Question

A retail enterprise operates two on-premises data centers that are 80 km apart. To support nightly data replication, they need at least 30 Gbps of highly available, low-latency private connectivity to a VPC in us-central1. The company wants to manage BGP on its own edge routers but does not want to lease long-haul fiber to reach a Google colocation facility that is 160 km away. Which solution delivers the required bandwidth and 99.99 % availability at the lowest recurring connectivity cost?

Order two Partner Interconnect circuits from different service providers and create redundant VLAN attachments that total at least 30 Gbps.
Use Cross-Cloud Interconnect to connect each data center to the VPC, placing ports in separate metro zones for redundancy.
Deploy two 100 Gbps Dedicated Interconnect links in the nearest Google colocation facility and aggregate them with Cloud Router.
Build four HA VPN tunnels over diverse ISP links from each data center, then use ECMP to aggregate the bandwidth.

Report Issue

Answer Description

Partner Interconnect lets customers reach Google's network through a supported service provider's point of presence, so the customer does not need to extend its network into a Google colocation facility. Each VLAN attachment can provide up to 50 Gbps. Creating redundant VLAN attachments that terminate on independent Partner Interconnect connections-preferably through different providers or diverse metro PoPs-meets Google's documented requirements for 99.99 % availability. Dedicated Interconnect would require leasing expensive fiber to the distant colocation site, while Cloud VPN would traverse the public internet and require many tunnels to achieve 30 Gbps with no SLA on latency. Cross-Cloud Interconnect is designed for cloud-to-cloud scenarios, not on-premises connectivity.

Ask Bash

Bash is our AI bot, trained to help you pass your exam. AI Generated Content may display inaccurate information, always double-check anything important.