A multinational logistics company is designing a hybrid network architecture to connect its primary on-premises data center with Google Cloud. The architecture must support continuous database replication requiring sustained bandwidth exceeding 12 Gbps with a guaranteed 99.99% availability SLA. Additionally, several workloads in Spoke VPC networks must communicate directly with one another without routing through a single point of failure. Which TWO architectural decisions must the cloud architect implement to meet these requirements?
- Provision Dedicated Interconnect circuits across two distinct edge availability domains in two separate metropolitan locations with redundant VLAN attachments.Cevap
- Establish direct VPC Network Peering connections between every pair of spoke VPCs that require inter-VPC communication.Cevap
- CDeploy multiple HA VPN tunnels over the public internet configured with active-active dynamic BGP routing to achieve 12 Gbps throughput with a 99.99% SLA.
- DConfigure a central Transit VPC peered with all spoke VPCs and rely on transitive routing over existing VPC Network Peering links to pass traffic between spoke VPCs.
Cevap
The architect must provision Dedicated Interconnect circuits across two distinct edge availability domains in two separate metropolitan locations, and establish direct VPC Network Peering connections between every pair of spoke VPCs requiring inter-communication.
To achieve a 99.99% SLA for bandwidth demands exceeding 10 Gbps, Google Cloud requires Dedicated Interconnect deployed across two metropolitan locations with dual circuits in each metro (4 connections total across 2 edge availability domains per metro). For inter-VPC traffic between spokes, because VPC Network Peering is non-transitive, spoke VPCs must be directly peered with each other to enable inter-spoke communication.
Adım Adım Çözüm
Anahtar Kavram
Hybrid Connectivity Selection & VPC Network Peering Non-Transitivity