GKE Multi-Cloud Kubernetes Deployment Skill

Overview

This skill enables the provisioning, management, and optimization of Google Kubernetes Engine (GKE) clusters with a consistent multi-cloud pattern. It follows the provision → connect → deploy pattern that can be extended to other providers (AWS EKS, Azure AKS).

Prerequisites

•Google Cloud SDK (gcloud) installed and authenticated
•kubectl installed
•Proper GCP authentication configured (gcloud auth login and gcloud auth application-default login)

When to Use This Skill

Use this skill when you need to:

•Provision GKE clusters on Google Cloud Platform
•Configure node pools and autoscaling for GKE clusters
•Connect kubectl to GKE clusters
•Apply cost optimization strategies to GKE deployments
•Deploy applications consistently across multiple cloud providers using a standardized pattern

Provision → Connect → Deploy Pattern

1. Provision Phase

•Create GKE clusters with appropriate configuration
•Set up node pools with desired specifications
•Configure autoscaling and network settings

2. Connect Phase

•Authenticate kubectl with the target cluster
•Verify cluster connectivity and readiness

3. Deploy Phase

•Deploy applications to the connected cluster
•Verify deployment status and health

GKE Cluster Provisioning

Basic Cluster Creation

bash

gcloud container clusters create CLUSTER_NAME \
  --zone=COMPUTE_ZONE \
  --num-nodes=NODE_COUNT \
  --machine-type=MACHINE_TYPE \
  --enable-autoscaling \
  --min-nodes=MIN_NODE_COUNT \
  --max-nodes=MAX_NODE_COUNT

Example:

bash

gcloud container clusters create my-cluster \
  --zone=us-central1-a \
  --num-nodes=3 \
  --machine-type=e2-medium \
  --enable-autoscaling \
  --min-nodes=1 \
  --max-nodes=10

Node Pool Management

Create additional node pools for specific workloads:

bash

gcloud container node-pools create POOL_NAME \
  --cluster=CLUSTER_NAME \
  --zone=COMPUTE_ZONE \
  --num-nodes=NODE_COUNT \
  --machine-type=MACHINE_TYPE \
  --enable-autoscaling \
  --min-nodes=MIN_NODE_COUNT \
  --max-nodes=MAX_NODE_COUNT

Example for GPU workloads:

bash

gcloud container node-pools create gpu-pool \
  --cluster=my-cluster \
  --zone=us-central1-a \
  --num-nodes=1 \
  --machine-type=n1-standard-4 \
  --accelerator=type=nvidia-tesla-k80,count=1 \
  --enable-autoscaling \
  --min-nodes=0 \
  --max-nodes=3

Connecting to the Cluster

After cluster creation, configure kubectl to connect to the cluster:

bash

gcloud container clusters get-credentials CLUSTER_NAME \
  --zone=COMPUTE_ZONE \
  --project=PROJECT_ID

Example:

bash

gcloud container clusters get-credentials my-cluster \
  --zone=us-central1-a \
  --project=my-project-id

Verify connection:

bash

kubectl cluster-info
kubectl get nodes

Cost Optimization Strategies

Machine Type Selection

Choose appropriate machine types based on workload requirements:

•e2-* series: General-purpose, cost-effective for most workloads
•n1-* series: Standard virtual machines for compute-intensive workloads
•n2-* and n2d-* series: Higher performance for compute-intensive workloads
•t2d-* and t2a-* series: Arm-based instances for cost-effective performance

Autoscaling Configuration

Configure cluster autoscaling:

bash

--enable-autoscaling \
--min-nodes=MIN_NODE_COUNT \
--max-nodes=MAX_NODE_COUNT \
--node-labels=workload-type=general-purpose

Configure vertical pod autoscaling:

bash

--enable-vertical-pod-autoscaling

Rightsizing Practices

•Monitor resource usage and adjust node pools accordingly
•Use preemptible nodes for fault-tolerant workloads
•Configure resource quotas and limits appropriately

Preemptible Nodes

For fault-tolerant workloads, use preemptible nodes for cost savings:

bash

gcloud container clusters create my-cluster \
  --zone=us-central1-a \
  --num-nodes=3 \
  --enable-autoscaling \
  --min-nodes=1 \
  --max-nodes=10 \
  --preemptible

Multi-Cloud Portability Patterns

The provision → connect → deploy pattern applies consistently across cloud providers:

GKE Pattern

•Provision: gcloud container clusters create ...
•Connect: gcloud container clusters get-credentials ...
•Deploy: Standard kubectl commands

Common Configuration Template

Store cloud-specific parameters in configuration files:

yaml

# gke-config.yaml
provider: gke
region: us-central1
zone: us-central1-a
node_count: 3
machine_type: e2-medium
min_nodes: 1
max_nodes: 10
disk_size_gb: 100
enable_autoscaling: true

Security Best Practices

Authentication

•Use Workload Identity for secure pod-to-GCP service communication
•Enable private clusters to restrict public access
•Configure master authorized networks for API server access

Example Secure Cluster

bash

gcloud container clusters create secure-cluster \
  --zone=us-central1-a \
  --num-nodes=1 \
  --machine-type=e2-medium \
  --enable-private-nodes \
  --master-ipv4-cidr-block=172.16.0.0/28 \
  --enable-master-authorized-networks \
  --master-authorized-networks=YOUR_IP/32

Monitoring and Maintenance

Health Checks

•Regularly check cluster status: gcloud container clusters describe CLUSTER_NAME
•Monitor node conditions: kubectl get nodes -o wide
•Review cluster events: kubectl get events

Updates

•Enable automatic node updates: --enable-autoupgrade
•Schedule maintenance windows appropriately

Troubleshooting Common Issues

Connection Issues

•Verify gcloud authentication: gcloud auth list
•Ensure kubectl is properly configured: kubectl config current-context
•Check if the cluster exists and is running: gcloud container clusters list

Resource Limitations

•Increase node pool sizes if pods are stuck in Pending state
•Adjust resource requests and limits appropriately
•Verify that quota is sufficient for the desired configuration

Extending to Other Providers

AWS EKS Pattern

•Provision: aws eks create-cluster ...
•Connect: aws eks update-kubeconfig ...
•Deploy: Standard kubectl commands

Azure AKS Pattern

•Provision: az aks create ...
•Connect: az aks get-credentials ...
•Deploy: Standard kubectl commands

This consistent pattern enables seamless multi-cloud Kubernetes deployments.