← Previous
NVIDIA AI-LLM Developers Certification Path
Next →
AI Models and LLM Development with NVIDIA
1. KCNA (Kubernetes and Cloud Native Associate)
Focuses on conceptual understanding rather than deep hands-on configuration.
- Format: MCQs
- Price: $250 (Euro 212)
- Duration: 1.5 hours
- Focus: Concepts over commands
- No heavy
kubectltroubleshooting
Syllabus
KCNA Syllabus Overview
| Domain | Weight | Key Areas |
|---|---|---|
| Kubernetes Fundamentals | 44% | Kubernetes Core ConceptsCluster Architecture (control plane & worker nodes)Administration basicsScheduling conceptsContainerization fundamentals |
| Container Orchestration | 28% | Networking model and CNI basicsSecurity fundamentalsTroubleshooting conceptsStorage (PV, PVC, StorageClass) |
| Cloud Native Application Delivery | 16% | Application delivery modelsCI/CD fundamentalsGitOps basicsDebugging concepts |
| Cloud Native Architecture | 12% | Observability (metrics, logs, traces)Cloud native ecosystem and CNCF landscapePrinciples (microservices, immutable infra, declarative systems)Community and collaboration fundamentals |
2. Certified Kubernetes Application Developer (CKAD)
Focuses on application development and deployment in Kubernetes.
- Format: Performance-based (hands-on tasks)
- price: $300 (Euro 254)
- Duration: 2 hours
- Focus: Application design, deployment, and maintenance
- Heavy use of
kubectlfor troubleshooting and configuration
CKAD Syllabus Overview
| Domain | Weight | Key Topics |
|---|---|---|
| Application Design and Build | 20% | Define, build, and modify container imagesChoose correct workload resource (Deployment, DaemonSet, Job, CronJob, StatefulSet)Multi-container Pod patterns (sidecar, init, ambassador, adapter)Persistent and ephemeral volumes |
| Application Deployment | 20% | Rolling updates and rollbacksBlue/Green and Canary strategies using Kubernetes primitivesDeploy applications with HelmManage configurations using Kustomize |
| Application Observability and Maintenance | 15% | API deprecations awarenessLiveness, Readiness, and Startup probeskubectl monitoring commandsContainer logsDebugging techniques |
| Application Environment, Configuration and Security | 25% | CRDs and OperatorsAuthentication, Authorization, Admission ControlRequests, Limits, ResourceQuotasConfigMapsSecrets (create and consume)ServiceAccountsSecurityContexts and Linux capabilities |
| Services and Networking | 20% | Service types (ClusterIP, NodePort, LoadBalancer, Headless)NetworkPolicies (ingress/egress rules)Ingress configuration and routingTroubleshooting service connectivity |
3 Certified Kubernetes Administrator (CKA)
Focuses on cluster administration and management.
- Format: Performance-based (hands-on tasks)
- price: $445 (Euro 254)
- Duration: 2 hours
- Focus: Cluster setup, maintenance, and troubleshooting
CKA Syllabus Overview
| Domain | Weight | Key Topics |
|---|---|---|
| Storage | 10% | Implement StorageClasses and dynamic provisioningConfigure volume types, access modes (RWO, ROX, RWX) and reclaim policiesCreate and manage PersistentVolumes (PV) and PersistentVolumeClaims (PVC) |
| Troubleshooting | 30% | Troubleshoot cluster and node failuresTroubleshoot control plane componentsMonitor cluster and application resource usageInspect and manage container logs and output streamsTroubleshoot Services and networking issues |
| Workloads & Scheduling | 15% | Perform rolling updates and rollbacksConfigure applications with ConfigMaps and SecretsImplement Horizontal Pod AutoscalingUnderstand self-healing primitives (ReplicaSet, Deployment)Configure scheduling constraints (resource limits, nodeSelector, affinity, taints/tolerations) |
| Cluster Architecture, Installation & Configuration | 25% | Manage RBAC (Roles, ClusterRoles, RoleBindings)Prepare infrastructure for cluster installationCreate and manage clusters using kubeadmUpgrade and manage cluster lifecycleConfigure highly available control planeInstall components using Helm and KustomizeUnderstand extension interfaces (CNI, CSI, CRI)Install and manage CRDs and Operators |
| Services & Networking | 20% | Pod-to-Pod connectivityImplement and enforce NetworkPoliciesUse Service types (ClusterIP, NodePort, LoadBalancer)Configure Gateway API and IngressUse Ingress controllers and resourcesUnderstand CoreDNS and service discovery |
