Deployment Environments

​

Three Deployment Contexts

​

Local Development

​

How It Works

$ applad up

1. Applad reads your config tree
2. Synthesizes a docker-compose.yml for the local environment
3. Starts services: database, storage, functions, messaging, realtime
4. Applies any pending migrations
5. Your API is live at http://localhost:8080

environments:
  - name: "development"
    url: "http://localhost:8080"
    infrastructure:
      type: "local"
      # Applad synthesizes docker-compose.yml and runs it locally
      # Same containers as staging and production
      # No authentication required to run locally

environment_overrides:
  development:
    primary:
      adapter: "sqlite"
      path: "./data/dev.db"
    analytics:
      adapter: "sqlite"
      path: "./data/analytics-dev.db"

​

Local Benefits

• Onboarding is applad up — any developer on any machine with Docker installed gets a full running stack immediately
• No runtime surprises on deploy — you’ve been running production containers locally the entire time
• Debugging is standard Docker tooling — docker logs, docker exec, docker compose ps
• Works offline — no cloud dependencies for development

​

VPS Deployment

• Full control over the entire stack
• Predictable costs (flat monthly rate)
• No vendor dependency at the infrastructure level
• Standard SSH access to debug anything

​

How It Works

environments:
  - name: "production"
    url: "https://api.myapp.com"
    infrastructure:
      type: "vps"
      host: "prod-01.acme-corp.com"
      user: "applad"
      ssh_key: "ci-github-actions"
      docker_host: "unix:///var/run/docker.sock"

​

VPS Sizing

​

VPS Providers

​

Cloud Providers On-Demand

• Run your core app on a Hetzner VPS (fast and cheap)
• Use S3 for storage (great pricing model)
• Use SES for high-volume email at scale
• Spin up a cloud compute instance for a heavy data processing job
• Use an AWS Mac instance for iOS builds — paying only for build duration, then tearing it down

​

Cloud as Utility, Not Platform

adapter: "s3"
config:
  bucket: ${S3_BUCKET}
  region: ${S3_REGION}

environment_overrides:
  development:
    adapter: "local"
    config:
      path: "./data/storage"

connections:
  - id: "primary"
    adapter: "postgres"
    url: ${DATABASE_URL}

environment_overrides:
  development:
    primary:
      adapter: "sqlite"
      path: "./data/dev.db"

email:
  provider: "resend"
  config:
    api_key: ${RESEND_API_KEY}
  environment_overrides:
    production:
      provider: "ses"
      config:
        access_key: ${AWS_SES_ACCESS_KEY}
        secret_key: ${AWS_SES_SECRET_KEY}

cloud_burst:
  enabled: true
  provider: "aws"
  instance_type: "t3.large"
  region: "eu-west-1"
  credentials: "aws-production"
  max_duration: 3600

​

Cloud Adapter Configuration

environments:
  - name: "production"
    infrastructure:
      type: "vps"
      host: "prod-01.acme-corp.com"
      # Core app runs on VPS

    # Cloud resources used on-demand alongside the VPS
    cloud_adapters:
      - provider: "aws"
        region: "eu-west-1"
        credentials: "aws-production"
        services:
          - s3
          - ses
          - rds

      - provider: "cloudflare"
        credentials: "cloudflare-prod"
        services:
          - r2
          - workers

​

The Continuum

Local dev (Docker Compose, SQLite)
  → VPS staging (Docker Compose, Postgres)
    → VPS production (Docker Compose, Postgres)
      → VPS + cloud storage adapter (S3/R2)
        → VPS + cloud database adapter (RDS)
          → Multi-VPS cluster
            → Kubernetes on cloud

​

Migration Example

environments:
  - name: "development"
    infrastructure:
      type: "local"

environments:
  - name: "development"
    infrastructure:
      type: "local"
  - name: "staging"
    infrastructure:
      type: "vps"
      host: "staging.myapp.com"

environments:
  - name: "production"
    infrastructure:
      type: "vps"
      host: "prod-01.myapp.com"
    cloud_adapters:
      - provider: "aws"
        services: [s3, ses]

​

Docker Compose Everywhere

​

Synthesis Example

connections:
  - id: "primary"
    adapter: "postgres"
    url: ${DATABASE_URL}

services:
  database:
    image: postgres:16-alpine
    environment:
      POSTGRES_DB: applad
      POSTGRES_USER: applad
      POSTGRES_PASSWORD: ${DATABASE_PASSWORD}
    volumes:
      - postgres-data:/var/lib/postgresql/data
    networks:
      - applad-network

​

Dry-Run and Diff

$ applad up --dry-run --diff

• The synthesized Docker Compose
• Every service that would start or restart
• Every config change
• Every migration pending

​

Environment Parity by Design

• Onboarding is applad up — any developer on any machine with Docker installed gets a full running stack immediately, no environment setup required beyond filling in .env
• No runtime surprises on deploy — you’ve been running production containers locally the entire time
• Debugging is standard Docker tooling — docker logs, docker exec, docker compose ps

​

Next Steps