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.
A company is deploying a large-scale Google Kubernetes Engine (GKE) cluster in a Workload VPC (vpc-workload) designed to scale up to 1,500 nodes. The networking architecture must satisfy the following requirements:
vpc-shared).vpc-workload via Cloud Interconnect to reach Pod IP addresses directly.Which GKE cluster networking mode and connectivity configuration should the company implement?
A VPC-native GKE cluster assigns Pod IP addresses from alias IP ranges backed by a subnet secondary IPv4 CIDR block. Instead of creating custom static routes for each node, Google Cloud programs Pod ranges directly into the Software-Defined Network (SDN) using automatically generated subnet routes.
vpc-workload, across VPC Network Peering to vpc-shared, and over Cloud Interconnect to on-premises networks without requiring NAT or IP masquerading.| Feature | VPC-Native Cluster | Routes-Based Cluster |
|---|---|---|
| Routing mechanism | Subnet routes via alias IPs | Per-node custom static routes |
| Route quota impact | Zero static routes consumed | 1 static route per node |
| Peering propagation | Automatic subnet exchange | Manual custom route export/import |
| VM anti-spoofing | Preserved and enforced | Disabled via IP forwarding |
Combining alias IP ranges, VPC Network Peering, and Cloud Router default subnet advertisements provides maximum scalability, strict anti-spoofing enforcement, line-rate SDN latency, and zero manual route maintenance.
This configuration pairs a routes-based GKE cluster (--no-enable-ip-alias) with default VPC Network Peering and Cloud Router behaviors, which only exchange and advertise VPC subnet routes automatically.
vpc-shared: In a routes-based cluster, Pod IP ranges (clusterIpv4Cidr) are implemented as custom static routes, not subnet routes. Default VPC Network Peering only exchanges subnet routes and will not propagate static routes unless custom route export and import are explicitly enabled.vpc-workload, heavily consuming VPC static route quotas.vpc-workload fails unless NAT/SNAT is introduced or custom routing is manually configured.The cluster were deployed in VPC-native mode using alias IP ranges from a subnet secondary range so that Pod addresses are backed by native subnet routes.
A routes-based GKE cluster allocates a /24 Pod IP address range to each cluster node and creates an individual VPC custom static route pointing to that node VM as the next hop. To share those routes beyond the local VPC, an administrator must explicitly enable custom route export and import on VPC Network Peering and configure custom route advertisements on Cloud Router.
The organization needed to integrate with an already-existing legacy routes-based cluster that remained well below VPC static route quotas and could tolerate manual custom route exchange.
This approach creates a VPC-native GKE cluster but interconnects the two VPC networks with HA VPN IPsec tunnels instead of VPC Network Peering, while configuring the IP masquerade agent to perform Source Network Address Translation (SNAT) from Pod IP addresses to node IP addresses.
vpc-shared and on-premises cannot observe original Pod source IPs, preventing Pod-level firewall rules and Pod-level network telemetry.The two VPC networks needed transitive routing through a third hub VPC or intentionally required hiding overlapping Pod IP addresses behind node IPs rather than allowing direct, NAT-free Pod communication.
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