Mastering Azure Kubernetes Service (AKS) Rapidly Build and Scale Your Containerized Applications with Microsoft Azure… (Abhishek Mishra)（Z-Library）

# Mastering Azure Kubernetes Service (AKS) — Reading Guide ## 【One-Line Pitch】 A hands-on, production-oriented guide for developers and DevOps engineers who want to master Azure Kubernetes Service — from container fundamentals through advanced topics like serverless Kubernetes, KEDA autoscaling, and Windows node pools. ## 【Book Arc】 - **Opening (~0%–9%)**: Establishes why containers and Kubernetes matter, contrasts VMs vs. containers, and maps the Azure container ecosystem (Container Registry, Container Instances, WebApp for Containers, AKS, Service Fabric). Sets up the managed-control-plane value proposition of AKS. - **Early (~9%–24%)**: Covers AKS architecture, core Kubernetes concepts (Pods, ReplicaSets, StatefulSets, DaemonSets, Namespaces), and foundational networking, storage, and scaling models — including Azure CNI, Persistent Volumes, and autoscalers. - **Early–Middle (~24%–39%)**: Moves into practical cluster operations: provisioning AKS via Portal/CLI/PowerShell, deploying containerized .NET Core apps, working with kubectl and Helm Charts, and managing clusters (scaling nodes/pods, configuring storage, upgrades, namespaces). - **Middle (~39%–52%)**: Deepens operational coverage with monitoring (Azure Monitor for Containers, kubelet logs), then pivots to advanced scenarios: serverless Kubernetes with Virtual Kubelet, event-driven autoscaling with KEDA, and running Windows containers on AKS. - **Late (~52%–end, per TOC)**: Covers CI/CD with Azure DevOps pipelines (containerize → push to ACR → deploy to AKS) and culminates in a near-production three-tier application walkthrough — designing the cluster, containerizing, and orchestrating end-to-end. ## 【Key Takeaways】 - **Containers are OS-level virtualization, not hardware-level** (Early): Unlike VMs that virtualize hardware via a hypervisor, containers share a single host OS through a container engine like Docker — making them lighter, faster to start, and cheaper to run at scale. - **AKS abstracts the Control Plane, not the nodes** (Early): Azure manages the complex control-plane infrastructure for you, but you retain full control over worker nodes — reducing operational overhead while keeping flexibility for node-level configuration. - **StatefulSets solve ordered, discoverable pod deployment** (Middle): When you need deterministic startup order (e.g., master before replica database) and stable network identities, ReplicaSets fall short because pods get dynamic IPs/names; StatefulSets assign fixed indexes and persistent names. - **Storage on AKS requires Persistent Volumes, not pod-local data** (Early): Pod data is ephemeral — if a pod dies, data dies. Azure Disks and Azure Files back Persistent Volumes, with Storage Classes and Persistent Volume Claims automating dynamic provisioning and attachment. - **Two autoscaling levers exist: nodes and pods** (Middle): The Cluster Autoscaler scales node count, while the Horizontal Pod Autoscaler scales pod replicas — and serverless nodes (via Azure Container Instance) enable rapid node scaling for spiky workloads. - **KEDA brings event-driven scaling to Kubernetes** (Early): Beyond CPU/memory-based HPA, KEDA scales pods based on external event sources (e.g., Azure Functions, queue lengths), enabling true serverless-style autoscaling on AKS. - **Virtual Kubelet extends AKS with serverless nodes** (Early): By registering Azure Container Instances as virtual nodes, you can burst beyond your node pool capacity without provisioning additional VMs — ideal for unpredictable or batch workloads. - **Helm and Azure DevOps streamline production deployments** (Early): Helm Charts package and version Kubernetes manifests for repeatable deployments, while Azure DevOps pipelines automate the containerize → push-to-ACR → deploy-to-AKS workflow. ## 【Reading Tips】 - **Skim Chapter 1's VM-vs-container comparison** (~9%–24%): If you already know Docker basics, the conceptual foundation is review — but the Azure container ecosystem mapping (ACR vs. ACI vs. WebApp vs. AKS vs. Service Fabric) is worth a careful read to understand service boundaries. - **Deep-read the storage and scaling chapters** (~24%–39%): Persistent Volumes, Storage Classes, and the two autoscalers are where most production issues arise. Work through the Azure Disk and Azure File examples hands-on rather than just reading. - **Treat the kubectl/Helm sections as reference material**: Don't memorize commands — bookmark the provisioning and deployment chapters (~39%–52%) and return to them when you need exact CLI syntax or YAML manifest patterns. - **Pay special attention to the StatefulSet vs. ReplicaSet discussion** (Middle): This is a common interview question and a real-world design decision — understand *why* ordering and stable identity matter, not just the syntax. - **The final three-tier application chapter is your capstone**: If you're short on time, read the architecture design decisions there even if you skip the step-by-step implementation — it ties together networking, storage, scaling, and DevOps in one coherent example. ## 【Coverage Limits】 This guide synthesizes the book's structure, core concepts, and operational themes from the available excerpts. Detailed step-by-step command syntax, specific YAML examples, and the complete three-tier application walkthrough are not fully covered here — consult the book's hands-on chapters for those. ##