sizing-profiles.yaml

# T-Shirt Sizing Profiles
#
# Overview:
# Predefined infrastructure configurations based on team size, workload requirements, and budget.
#
# Sizing Matrix:
# | Profile | Team Size | Workloads | Use Case |
# |---------|-----------|-----------|----------|
# | Small   | < 10      | Dev/POC   | Proof of concept, development |
# | Medium  | 10-50     | Standard  | Standard production |
# | Large   | 50-200    | Enterprise| Enterprise production |
# | XLarge  | 200+      | Critical  | Mission critical, multi-region |

profiles:
  small:
    profile: small
    description: "Development, POC, Small Teams (< 10 developers)"

    infrastructure:
      aks:
        name: "${PROJECT}-${ENV}-aks"
        kubernetes_version: "1.30"
        node_pool:
          name: "system"
          node_count: 3
          vm_size: "Standard_D2s_v5"
          os_disk_size_gb: 50
          max_pods: 30
        auto_scaling:
          enabled: false
        features:
          azure_policy: false
          oidc_issuer: true
          workload_identity: true

      acr:
        name: "${PROJECT}${ENV}acr"
        sku: "Basic"
        geo_replication: false

      keyvault:
        name: "${PROJECT}-${ENV}-kv"
        sku: "standard"

      networking:
        vnet_cidr: "10.0.0.0/16"
        aks_subnet: "10.0.0.0/23" # /23 = 512 IPs (enough for 3 nodes)

    databases:
      postgresql:
        enabled: true
        name: "${PROJECT}-${ENV}-postgres"
        version: "16"
        sku: "Standard_B1ms"
        storage_gb: 32
        ha_enabled: false
        backup_retention_days: 7

      redis:
        enabled: true
        name: "${PROJECT}-${ENV}-redis"
        sku: "Basic"
        family: "C"
        capacity: 0

    ai_foundry:
      enabled: false # Optional for Small
      models:
        gpt4o:
          capacity_tpm: 10000
        embedding:
          capacity_tpm: 30000

    observability:
      log_analytics:
        retention_days: 30
      prometheus:
        retention_days: 7
        storage_gb: 10

  medium:
    profile: medium
    description: "Standard Production (10-50 developers)"

    infrastructure:
      aks:
        name: "${PROJECT}-${ENV}-aks"
        kubernetes_version: "1.30"
        node_pool:
          name: "system"
          node_count: 5
          vm_size: "Standard_D4s_v5"
          os_disk_size_gb: 100
          max_pods: 50
        auto_scaling:
          enabled: true
          min_nodes: 3
          max_nodes: 10
        features:
          azure_policy: true
          oidc_issuer: true
          workload_identity: true
          azure_monitor: true
          defender: true

      acr:
        name: "${PROJECT}${ENV}acr"
        sku: "Standard"
        geo_replication: false
        retention_days: 30

      keyvault:
        name: "${PROJECT}-${ENV}-kv"
        sku: "standard"
        soft_delete_days: 90

      networking:
        vnet_cidr: "10.0.0.0/16"
        aks_subnet: "10.0.0.0/22" # /22 = 1024 IPs
        services_subnet: "10.0.4.0/24"
        private_endpoints_subnet: "10.0.5.0/24"
        private_cluster: false

    databases:
      postgresql:
        enabled: true
        name: "${PROJECT}-${ENV}-postgres"
        version: "16"
        sku: "Standard_D2ds_v5"
        storage_gb: 128
        ha_enabled: false
        backup_retention_days: 14
        geo_redundant_backup: false

      redis:
        enabled: true
        name: "${PROJECT}-${ENV}-redis"
        sku: "Standard"
        family: "C"
        capacity: 1

    ai_foundry:
      enabled: true
      models:
        gpt4o:
          capacity_tpm: 50000
        gpt4o_mini:
          capacity_tpm: 100000
        embedding:
          capacity_tpm: 100000
      ai_search:
        sku: "basic"
        replicas: 1
        partitions: 1

    observability:
      log_analytics:
        retention_days: 60
      prometheus:
        retention_days: 15
        storage_gb: 50
      grafana:
        sku: "Standard"

  large:
    profile: large
    description: "Enterprise Production (50-200 developers)"

    infrastructure:
      aks:
        name: "${PROJECT}-${ENV}-aks"
        kubernetes_version: "1.30"
        node_pools:
          - name: "system"
            node_count: 3
            vm_size: "Standard_D4s_v5"
            os_disk_size_gb: 100
            mode: "System"

          - name: "workloads"
            node_count: 5
            vm_size: "Standard_D8s_v5"
            os_disk_size_gb: 200
            max_pods: 100
            mode: "User"
            auto_scaling:
              enabled: true
              min_nodes: 3
              max_nodes: 15

          - name: "ai"
            node_count: 2
            vm_size: "Standard_NC6s_v3" # GPU
            os_disk_size_gb: 200
            max_pods: 30
            mode: "User"
            taints:
              - "workload=ai:NoSchedule"

        features:
          azure_policy: true
          oidc_issuer: true
          workload_identity: true
          azure_monitor: true
          defender: true
          private_cluster: true

      acr:
        name: "${PROJECT}${ENV}acr"
        sku: "Premium"
        geo_replication: true
        replications:
          - location: "eastus"
        retention_days: 90
        zone_redundancy: true

      keyvault:
        name: "${PROJECT}-${ENV}-kv"
        sku: "premium" # HSM support
        soft_delete_days: 90
        purge_protection: true

      networking:
        vnet_cidr: "10.0.0.0/14" # Large CIDR
        aks_subnet: "10.0.0.0/20" # 4096 IPs
        services_subnet: "10.0.16.0/22"
        private_endpoints_subnet: "10.0.20.0/24"
        appgw_subnet: "10.0.21.0/24"
        private_cluster: true
        application_gateway:
          enabled: true
          sku: "WAF_v2"
          capacity: 2

    databases:
      postgresql:
        enabled: true
        name: "${PROJECT}-${ENV}-postgres"
        version: "16"
        sku: "Standard_D4ds_v5"
        storage_gb: 256
        ha_enabled: true
        ha_mode: "ZoneRedundant"
        backup_retention_days: 30
        geo_redundant_backup: true

      redis:
        enabled: true
        name: "${PROJECT}-${ENV}-redis"
        sku: "Premium"
        family: "P"
        capacity: 1
        zones: ["1", "2", "3"]

    ai_foundry:
      enabled: true
      models:
        gpt4o:
          capacity_tpm: 150000
        gpt4o_mini:
          capacity_tpm: 300000
        embedding:
          capacity_tpm: 300000
      ai_search:
        sku: "standard"
        replicas: 2
        partitions: 1
      content_safety:
        enabled: true

    observability:
      log_analytics:
        retention_days: 90
      prometheus:
        retention_days: 30
        storage_gb: 100
        ha_enabled: true
      grafana:
        sku: "Standard"
      alertmanager:
        pagerduty: true

  xlarge:
    profile: xlarge
    description: "Mission Critical (200+ developers, multi-region)"
    multi_region: true

    regions:
      primary: "centralus"
      secondary: "eastus"

    infrastructure:
      primary:
        aks:
          name: "${PROJECT}-${ENV}-aks-primary"
          kubernetes_version: "1.30"
          node_pools:
            - name: "system"
              node_count: 5
              vm_size: "Standard_D8s_v5"
              zones: ["1", "2", "3"]

            - name: "workloads"
              node_count: 10
              vm_size: "Standard_D16s_v5"
              zones: ["1", "2", "3"]
              auto_scaling:
                enabled: true
                min_nodes: 5
                max_nodes: 30

            - name: "ai"
              node_count: 4
              vm_size: "Standard_NC12s_v3"
              zones: ["1", "2"]

          features:
            azure_policy: true
            private_cluster: true
            defender: true
            uptime_sla: true # 99.95% SLA

      secondary:
        aks:
          name: "${PROJECT}-${ENV}-aks-secondary"
          node_pools:
            - name: "system"
              node_count: 3
              vm_size: "Standard_D4s_v5"

            - name: "workloads"
              node_count: 5
              vm_size: "Standard_D8s_v5"
              auto_scaling:
                enabled: true
                min_nodes: 3
                max_nodes: 15

      acr:
        name: "${PROJECT}${ENV}acr"
        sku: "Premium"
        geo_replication: true
        replications:
          - location: "eastus"
            zone_redundancy: true
          - location: "westeurope"
            zone_redundancy: true

      front_door:
        enabled: true
        sku: "Premium"
        waf_policy: true

    databases:
      postgresql:
        primary:
          name: "${PROJECT}-${ENV}-postgres"
          sku: "Standard_D8ds_v5"
          storage_gb: 512
          ha_enabled: true
          ha_mode: "ZoneRedundant"
        read_replicas:
          - region: "eastus"
            sku: "Standard_D4ds_v5"

      redis:
        sku: "Premium"
        family: "P"
        capacity: 3
        geo_replication: true

      cosmos_db:
        enabled: true
        api: "sql"
        multi_region_writes: true
        regions:
          - "brazilsouth"
          - "eastus"

    ai_foundry:
      multi_region: true
      regions:
        primary:
          models:
            gpt4o:
              capacity_tpm: 300000
            embedding:
              capacity_tpm: 500000
        secondary:
          models:
            gpt4o:
              capacity_tpm: 150000
            embedding:
              capacity_tpm: 250000
      ai_search:
        sku: "standard2"
        replicas: 3
        partitions: 2

    disaster_recovery:
      rpo_minutes: 15
      rto_minutes: 60
      backup:
        cross_region: true
        retention_days: 90

metadata:
  version: "1.0.0"
  usage_in_issue_templates:
    - type: dropdown
      id: sizing
      attributes:
        label: T-Shirt Size
        description: Select infrastructure size based on your needs
        options:
          - "🟢 Small - Dev/POC"
          - "🟡 Medium - Standard Production"
          - "🔴 Large - Enterprise"
          - "⚫ XLarge - Mission Critical"