Skip to content

Microsoft Azure

Cloud Adoption Framework

Fundamental concepts

everyone must read and understand the concepts/vocabulary described in:

Fundamental decisions

How to lead the portfolio:

  • Innovation / development-led: 1 workload = 1 landing zone = 1 environment, focus on cloud adoption/automation, environments prod/non-prod
  • Operations-led: larger organization and ops team, landing zones group application into archetypes,
  • Migration-led: datacenter migration, focus on assets (not always with associated workload)
  • Governance-led: idk

Which naming convention: (see tutorial)

a naming convention must fulfill 2 requirements

  • business side: identify the teams involved and business owners
  • operational side: which workload, application, environment, criticality



  • subscription_id
  • deployment region

Resource group

  • Included Resources
  • subscription id

Management groups

  • Included subscriptions

Subscription (organization wide)

  • Azure account id
  • Azure AD instance (tenant)
  • Azure offer id (list here)
  • Legal agreement
  • Scale limit

Azure account

  • Email
  • Contact information
  • Billing details
  • Billing method
  • Azure Role (Service Administrator if this user created the subscription)

workload: collection of assets (Resources) that delivers business value (servers, applications, data, devices). 1 workload = 1 resource group

portfolio: collection of workloads

landing zone: remember there can be MULTIPLE landing zones. 1 landing zone = 1 subscription

platform foundation: provide foundational, shared utilities tailored to one or more workloads

cloud platform: different platform foundations (and deployed workloads) / can be multicloud platform. 1 cloud platform = 1 management group node

Linker un cluster AKS à kubectl

  • Créer son cluster en cliquant partout
  • Récuperer le resource group du cluster:
az resource list --name 'mon-super-cluster' | grep "\"resourceGroup\":"
  • Remplir le kubeconfig avec cette commande get:
az aks get-credentials --name <clusterName> --resource-group <resourceGroup>
  • On peut maintenant utiliser kubectl !