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 VPC-native Google Kubernetes Engine (GKE) cluster in us-central1 hosting internal HTTPS and gRPC microservices. You must expose these services with the following requirements:
us-central1, europe-west1, and an on-premises network connected via Cloud Interconnect in us-east4 must connect over a private RFC 1918 IP address.GCE_VM_IP_PORT Network Endpoint Groups (NEGs).Which GKE load balancing implementation should you deploy?
External GKE Ingress (configured with kubernetes.io/ingress.class: "gce" or by omitting the class annotation) deploys a Classic External Application Load Balancer backed by globally distributed Google Front End (GFE) proxies. It is designed specifically for internet-facing applications requiring a publicly routable global Anycast IP address and container-native GCE_VM_IP_PORT NEGs.
us-central1, europe-west1, and on-premises data centers would have to route requests out to a public IP endpoint (requiring external NAT or public internet routing) rather than traversing private internal VPC routing and private Cloud Interconnect VLAN attachments.When designing GKE networking, strictly separate north-south internet-facing traffic (served by External Ingress, External Gateway, or External LoadBalancer Services) from internal service-to-service and hybrid traffic (served by Internal Gateway, Internal Ingress, or Internal LoadBalancer Services).
Internal GKE Ingress uses the built-in GKE Ingress controller (triggered by the kubernetes.io/ingress.class: "gce-internal" annotation) to provision a regional internal Application Load Balancer. It deploys managed Envoy proxies inside a dedicated regional proxy-only subnet within your VPC network and routes Layer 7 HTTP(S) traffic directly to Pods via container-native GCE_VM_IP_PORT NEGs.
us-central1) and VPC (or peering and hybrid links terminating in us-central1).europe-west1 and on-premises traffic arriving over Cloud Interconnect VLAN attachments in us-east4 will fail to connect to the regional internal VIP.INTERNAL_MANAGED) locked to the cluster's region.us-central1 region.When you need Layer 7 internal load balancing in GKE with cross-region client reachability (global access), migrating from GKE Ingress to an internal GKE Gateway exposes the GatewayClass options required for global access and cross-region internal Application Load Balancers.
An internal GKE Gateway uses the GKE Gateway controller (the Kubernetes Gateway API implementation for GKE) to provision an Envoy-based internal Application Load Balancer (INTERNAL_MANAGED). It separates infrastructure entry-point definitions (Gateway and GatewayClass) from application routing rules (HTTPRoute), delivering Layer 7 HTTP, HTTPS, and HTTP/2 load balancing inside a VPC network.
GatewayClass that supports global access (or cross-region internal mode) and allocating an internal IP from a VPC subnet, clients in us-central1, europe-west1, and on-premises networks arriving via Cloud Interconnect in us-east4 can all reach the private RFC 1918 VIP.Gateway resource terminates TLS using self-managed certificates at the Envoy proxy tier (deployed in a regional proxy-only subnet), while HTTPRoute resources evaluate URL paths and forward requests to the appropriate backend Kubernetes Services.GCE_VM_IP_PORT zonal NEGs, routing traffic from the Envoy proxies directly to healthy Pod IP addresses without extra node-level kube-proxy hops.GCE_VM_IP_PORT NEGs.Unlike internal GKE Ingress—which prohibits enabling global access—and internal LoadBalancer Services—which lack Layer 7 URL routing, TLS termination, and GCE_VM_IP_PORT NEGs—an internal GKE Gateway is the only Kubernetes-native GKE construct that simultaneously satisfies cross-region private VPC reachability, Layer 7 routing, and direct-to-Pod load balancing.
An internal LoadBalancer Service (configured with spec.loadBalancerClass: "networking.gke.io/l4-regional-internal" or the networking.gke.io/load-balancer-type: "Internal" annotation) instructs GKE to provision a Google Cloud internal passthrough Network Load Balancer. Built on the Andromeda network virtualization stack, it operates at Layer 4 to distribute TCP or UDP traffic within a private VPC network.
LoadBalancer Services use node-level GCE_VM_IP NEGs (when GKE subsetting is enabled) or unmanaged instance groups, rather than Pod-level GCE_VM_IP_PORT NEGs. Traffic arrives at a cluster node first and may require an extra intra-cluster hop via kube-proxy or eBPF.europe-west1 and on-premises networks in us-east4 to reach the internal VIP.In multi-tier GKE architectures, internal LoadBalancer Services are ideal for backend data tiers (such as databases or message brokers), whereas Layer 7 microservice routing requires an internal Application Load Balancer managed via Gateway or Ingress.
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