Getting Started with Google Kubernetes Engine (Nilanjana Saha, Nishanta Banik)（Z-Library）

【One-Line Pitch】 A hands-on, console-and-CLI walkthrough of Google Kubernetes Engine that takes you from a blank GCP account to running, securing, monitoring, and cost-tuning containerized workloads. Best for developers, DevOps engineers, and cloud newcomers who want practical GKE fluency rather than Kubernetes theory. 【Book Arc】 - **Opening (~0%–12%)**: Orients you in Google Cloud fundamentals — account setup, compute/storage/database options, and where GKE fits as a managed Kubernetes service — so you understand the platform before touching a cluster. - **Early (~12%–31%)**: Builds your first cluster: prerequisites, Console and Terraform creation paths, Autopilot vs. Standard modes, and the toolchain (gcloud CLI, kubectl, K9s, OpenLens, Helm, Argo CD, Prometheus/Grafana). - **Middle (~31%–54%)**: Moves into deployment and networking — Kustomize vs. Helm for manifest management, Service types, internal/external communication, load balancing, network policies, DNS, and Ingress. - **Late (~54%–70%)**: Shifts to operations — Logs Explorer, PromQL-based alerting, monitoring and troubleshooting workflows, and Istio for traffic management, mTLS, and authorization. - **Ending (~70%–100%)**: Closes on cost optimization — storage, network, and workload-level controls, image management, and broader GCP cost practices, reinforced with scenario-based exercises and MCQs. 【Key Takeaways】 - **GKE is a managed layer over Kubernetes, not a replacement** (Early): the book stresses the managed-vs-open-source distinction — Google handles node config, autoscaling, upgrades, and baseline security, freeing you to focus on apps. - **Autopilot vs. Standard is the first real design decision** (Early): Autopilot offloads operational overhead; Standard gives you node-pool and upgrade control. The trade-off shapes everything downstream. - **Tooling fluency matters as much as cluster creation** (Early): gcloud, kubectl, Kustomize, Helm, Argo CD, and the Prometheus/Grafana stack are treated as core skills, not extras. - **Kustomize and Helm solve different problems** (Middle): Kustomize suits lightweight YAML overlays for environment-specific configs; Helm provides templating and package management. The book frames them as complementary, not competing. - **Service type choice drives connectivity** (Middle): ClusterIP for internal traffic, NodePort for dev/testing, LoadBalancer for internet-facing apps — with global vs. regional load balancing as a further decision. - **Security is layered, not bolted on** (Early/Middle): vulnerability scanning, fine-grained IAM, network policies, and Istio's mTLS and authorization policies form a defense-in-depth story. - **Observability is a workflow, not a dashboard** (Late): Logs Explorer filtering, PromQL alert rules, and a monitor-analyze-adjust cycle for load balancers show operations as iterative. - **Cost is an engineering concern** (Ending): storage, network, and workload-level optimizations plus image management are presented as ongoing practices, not afterthoughts. 【Reading Tips】 - **Deep-read the cluster-creation and tooling chapters** (Early): these are the foundation; skimming them makes later networking and Istio sections harder to follow. - **Skim the GCP fundamentals if you already know cloud basics** (Opening): useful for newcomers, but experienced readers can move quickly to GKE-specific content. - **Treat the scenario/MCQ sections as self-checks** (throughout): they reinforce decisions like Kustomize vs. Helm and Service type selection. - **Pair CLI commands with the Console walkthroughs**: the book alternates both, and following along in a real project cements the workflow. - **Watch for cost warnings**: the text repeatedly notes that clusters and scanning incur charges — clean up resources after exercises. 【Coverage Limits】 This guide is synthesized from stratified excerpts covering roughly the first half of the book in detail, with later chapters (Istio, cost optimization) represented mainly through table-of-contents and partial material. Specific figures, full code listings, and chapter-level nuance beyond the excerpts are not covered.