Prepare and test your skills

Prepare and test your skills

Free sample question with answers and detailed explanation below. The Professional Cloud Network Engineer practice bank has an additional 2,437 practice questions.Try 10 free realistic questions: instant score, no credit card, and no account required.
Your organization runs a hybrid workload in a single VPC network with global dynamic routing mode enabled across us-central1 and us-east4. To achieve a 99.99% availability SLA, you deployed four 10-Gbps Dedicated Interconnect connections:
us-central1: Two connections in separate edge availability domains, each with a 10-Gbps VLAN attachment.us-east4: Two connections in separate edge availability domains, each with a 10-Gbps VLAN attachment.During peak traffic, us-central1 sends 14 Gbps on-premises. When one us-central1 link enters maintenance, the remaining 10-Gbps link saturates and drops packets without spilling to us-east4, and operations lacks congestion alerts.
Which strategy should you implement to resolve this capacity bottleneck, preserve optimal routing, and detect performance degradation?
us-central1 with BGP Bidirectional Forwarding Detection (BFD) enabled on the VPN tunnels for faster failover, and configure hot-potato routing inside Google Cloud so excess traffic exits early via us-east4.This design wraps traffic inside HA VPN over Cloud Interconnect IPsec tunnels, attempts to use Bidirectional Forwarding Detection (BFD) on the VPN BGP sessions, and applies hot-potato routing within the Google Cloud VPC.
End-to-end IPsec encryption was the primary requirement on already properly sized physical Interconnect links, BFD was enabled on the underlying Interconnect BGP sessions rather than VPN tunnels, and cold-potato routing was preserved inside Google Cloud.
This strategy aligns physical and logical Cloud Interconnect capacity with Cloud Router route-selection behavior in a multi-region VPC, while pairing deterministic cold-potato routing with continuous network observability.
us-east4) if every VLAN attachment in us-central1 goes down. Upgrading each of the four connections and VLAN attachments across both edge availability domains (EADs) (for example, using LACP-aggregated links up to 200 Gbps per path) ensures total metro bandwidth is at least twice the attachment demand and every individual link can carry the full 14-Gbps load during maintenance.us-east4 if all us-central1 attachments fail.Because Cloud Router will not load-balance across regions when local attachments are merely congested, provisioning sufficient per-EAD physical and VLAN attachment capacity—combined with active-active flow-based ECMP and Cloud Monitoring alerts—is the only design that prevents maintenance-induced drops while preserving low-latency local egress.
This configuration restricts dynamic route propagation to individual regions, alters Equal Cost Multi-Path (ECMP) hashing from flow-based to per-packet distribution on the on-premises routers, and depends on Network Analyzer for outage troubleshooting.
us-central1 connections fail, us-central1 workloads can no longer fail over across us-east4 Interconnect links, invalidating the 99.99% availability SLA.us-central1, so 14 Gbps of egress traffic still saturates a single 10-Gbps link during maintenance.Strict data residency rules required keeping all traffic inside a single region (accepting a 99.9% SLA), flow-based hashing was used instead of per-packet ECMP, and Cloud Monitoring replaced Network Analyzer for real-time congestion alerting.
This approach raises the logical VLAN attachment limit above the underlying 10-Gbps physical port speed, changes the intra-metro topology from active-active to active-passive, and relies on Network Topology for operational visibility.
us-central1 traffic onto a single 10-Gbps link even during normal operations, leaving the second 10-Gbps link idle and causing continuous packet loss.Physical connections were already upgraded to 20 Gbps or higher per EAD, stateful security devices strictly mandated active-passive failover, and Cloud Monitoring alerting policies were configured alongside Network Topology.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated Validated