---
title: "Interview, certification, and glossary"
chapter: "17"
---

# Interview, certification, and glossary

Use **CLUSTER** to answer design questions:

- **C**ontext: outcome, users, data, scale, existing systems.
- **L**evels of reliability: SLO, RTO, RPO, failure scope.
- **U**sers and identities: IAM, RBAC, workload identity, tenants.
- **S**ervices and state: workloads, traffic, storage, dependencies.
- **T**rust and threats: supply chain, network, policy, secrets.
- **E**lasticity and economics: requests, scaling, quota, unit cost.
- **R**ollout and recovery: delivery, observability, upgrades, DR.

## Professional Cloud Architect checkpoint

Google's standard Professional Cloud Architect exam is two hours with 50–60
multiple-choice and multiple-select questions. It includes two case studies;
case-study questions make up 20–30%. This site uses one 60-question original
GKE-heavy mock and two fictional case studies. It is practice, not copied exam
content.

Official source: [Professional Cloud Architect certification](https://cloud.google.com/learn/certification/cloud-architect).

## Compact glossary

| Term | Plain meaning |
|---|---|
| Autopilot | GKE mode where Google manages node infrastructure and many defaults |
| Standard | GKE mode with customer-managed node pools and more control |
| Cluster | Kubernetes control plane plus registered compute |
| Node | Worker machine that runs Pods |
| Pod | Smallest Kubernetes scheduled unit |
| Deployment | Reconciles and rolls out stateless Pods |
| StatefulSet | Workload with stable identity and ordered behavior |
| Service | Stable discovery and virtual endpoint for Pods |
| Gateway API | Kubernetes traffic-routing API with role separation |
| HPA | Changes Pod replica count from metrics |
| VPA | Recommends or changes Pod resource requests |
| PDB | Limits voluntary concurrent Pod disruption |
| RBAC | Kubernetes role-based authorization |
| IAM | Google Cloud authorization |
| WIF | Workload Identity Federation; keyless workload access |
| CSI | Container Storage Interface |
| CNI | Container Network Interface |
| SLO | Measurable reliability target |
| RTO | Maximum acceptable recovery time |
| RPO | Maximum acceptable data loss measured in time |

## Official references

- [GKE documentation](https://cloud.google.com/kubernetes-engine/docs)
- [GKE best practices](https://cloud.google.com/kubernetes-engine/docs/best-practices)
- [Google Cloud Well-Architected Framework](https://cloud.google.com/architecture/framework)
- [GKE security best practices](https://cloud.google.com/docs/security/gke-security-bps)
- [Autopilot and Standard comparison](https://cloud.google.com/kubernetes-engine/docs/resources/autopilot-standard-feature-comparison)
- [GKE fleets](https://cloud.google.com/kubernetes-engine/fleet-management/docs/fleet-concepts)

Features, versions, limits, and certification formats change. Verify production
details in current official documentation.