Skip to content

Series Portal

Once an architecture decision has narrowed the workload to a specific Azure service, use this page to jump directly to the sibling deep-guide for that service. This is a router, not a selection matrix — for service-selection guidance, start with the Architecture Decision Matrix, one of the workload cheatsheets, or the scenario router in Start Here.

Each entry links to a companion practical guide that follows the same core structure documented in the Series Lab Contract: Start Here → Platform → Best Practices → Operations → Troubleshooting → Reference.

Still choosing a service? Start with the matrix

This portal assumes the service is already chosen. If you are still deciding which Azure service fits the workload, start with the Architecture Decision Matrix — it maps workload types to service combinations and hands off to the same sibling deep-guides listed below.

When to use this portal

  • You already know which Azure service the workload will run on. [Documented]
  • You need production-grade platform, best-practices, or troubleshooting depth beyond what a decision matrix or cheatsheet provides. [Inferred]
  • You are cross-referencing a multi-service architecture (for example, App Service front + Functions back + Storage) and want each service's deep guide side by side. [Correlated]

Sibling deep-guides

Each row links to the deployed deep-guide site plus its two canonical entry points: Learning Paths (role-based reading paths) and Scenario Router (symptom or workload-first routing). Both entry points follow the series-wide Start Here shape rolled out across the sibling repositories. [Documented]

Repository Focus area When to use Key entry points
Azure Virtual Machines Compute — IaaS, custom OS, lift-and-shift Workload requires OS-level control, custom kernel, legacy binaries, or GPU workloads. [Documented] Learning Paths · Scenario Router
Azure Networking Connectivity — VNet, Private Link, Front Door, DNS Designing hub-spoke or Virtual WAN topology, or troubleshooting hybrid connectivity. [Documented] Learning Paths · Scenario Router
Azure Storage Data at rest — Blob, Files, Tables, Queues Selecting storage tier, replicating across regions, or securing with Private Endpoints. [Documented] Learning Paths · Scenario Router
Azure App Service Compute — managed PaaS for HTTP workloads Hosting web apps, APIs, or Windows/Linux runtimes with managed platform. [Documented] Learning Paths · Scenario Router
Azure Functions Compute — event-driven serverless Bursty event processing, timers, webhooks, or lightweight integration glue. [Documented] Learning Paths · Scenario Router pending — check Start Here
Azure Communication Services Communications — chat, voice, video, SMS Embedding real-time communications into apps or building contact-center flows. [Documented] Learning Paths · Scenario Router
Azure Container Apps Compute — managed container runtime Running containers without operating a full Kubernetes cluster; KEDA-driven scale to zero. [Documented] Learning Paths · Scenario Router
Azure Kubernetes Service (AKS) Compute — managed Kubernetes Full Kubernetes control, service mesh, operators, or platform-team-owned clusters. [Documented] Learning Paths · Scenario Router
Azure Monitoring Observability — Log Analytics, Metrics, alerts Setting up diagnostics, KQL queries, or cross-service telemetry pipelines. [Documented] Learning Paths · Scenario Router
Azure Architecture This guide — architecture design, reviews, patterns Architecture-level decisions, workload blueprints, Well-Architected reviews, design labs. [Documented] Learning Paths · Scenario Router

Route from architecture decisions to service depth

Use this pattern when moving from an architecture cheatsheet or decision matrix into a service-specific deep guide.

flowchart TD
    A[Architecture decision made] --> B{Which service surface?}
    B -->|Compute PaaS HTTP| C[App Service guide]
    B -->|Compute serverless events| D[Functions guide]
    B -->|Compute containers managed| E[Container Apps guide]
    B -->|Compute containers Kubernetes| F[AKS guide]
    B -->|Compute IaaS OS control| G[Virtual Machines guide]
    B -->|Data at rest| H[Storage guide]
    B -->|Connectivity and DNS| I[Networking guide]
    B -->|Real-time comms| J[Communication Services guide]
    B -->|Observability| K[Monitoring guide]
    C --> L[Read Learning Paths and Scenario Router]
    D --> L
    E --> L
    F --> L
    G --> L
    H --> L
    I --> L
    J --> L
    K --> L
    L --> M[Deep-guide platform + best practices + troubleshooting]

Selection notes

  • The Container Apps and AKS guides both cover container workloads but target different ownership models. Choose Container Apps when the workload does not need cluster-level customization; choose AKS when the platform team owns operator-grade control. [Correlated]
  • The App Service and Functions guides frequently pair for HTTP + background patterns. Their Learning Paths cross-reference each other's runtime and hosting concepts. [Observed]
  • The Networking and Storage guides are prerequisites for most workload deep-guides because Private Endpoints, DNS, and managed identities are pervasive. [Inferred]
  • The Monitoring guide is horizontally applicable to every workload; its KQL query packs are the fastest way to build diagnostic dashboards for a running service. [Documented]
  • Every sibling deep-guide follows the same top-level information architecture (Start Here → Platform → Best Practices → Operations → Troubleshooting → Reference), so once you learn one guide's navigation, the others feel familiar. [Documented]

Notes on entry-point stability

  • Learning Paths and Scenario Router pages exist under every sibling repository's docs/start-here/ and are rendered on the deployed MkDocs Material site at /start-here/learning-paths/ and /start-here/scenario-router/ respectively. [Documented]
  • The Azure Functions guide's Scenario Router is intentionally deferred — see Series Lab Contract for scope details. Its Learning Paths page is live. [Documented]
  • If any linked entry-point URL returns a 404, browse to the deep-guide home in that row's first column and use the site navigation to locate Learning Paths or Scenario Router. [Documented]

See Also

Sources

  • https://learn.microsoft.com/en-us/azure/architecture/
  • https://learn.microsoft.com/en-us/azure/architecture/guide/technology-choices/