Architecture Diagram — Hobby / MVP
Component Name
COMPONENT TYPEWhat It Does
Description here
Cloud Services
Service A, Service BEst. Monthly Cost
$0-10
Why At This Scale
Reason here
Infrastructure as Code — Terraform (Hobby)
# Cloudflare Workers + D1 Database
# Perfect for hobby projects with minimal cost
terraform {
required_providers {
cloudflare = {
source = "cloudflare/cloudflare"
version = "~> 4.0"
}
}
}
provider "cloudflare" {
api_token = var.cloudflare_api_token
account_id = var.cloudflare_account_id
}
# D1 Database (Serverless SQLite)
resource "cloudflare_d1_database" "app_db" {
account_id = var.cloudflare_account_id
name = "observatory-hobby-db"
}
# Worker Script
resource "cloudflare_worker_script" "api" {
name = "observatory-api"
content = file("./worker.js")
d1_database {
binding = "DB"
database_id = cloudflare_d1_database.app_db.id
}
}
# Custom Domain Routing
resource "cloudflare_worker_route" "api_route" {
pattern = "api.example.com/*"
worker_name = cloudflare_worker_script.api.name
zone_id = var.zone_id
}
Click to Learn More
Line 24-28
D1 Database Binding — Connects the Worker to a serverless SQLite database. No connection pooling needed.
Line 31-40
Worker Script — Edge-computed API running globally. Cold starts under 50ms.
Line 43-48
Route Pattern — Maps custom domain paths to the Worker. Handles SSL automatically.
Architecture Diagram — Startup
Component Name
COMPONENT TYPEWhat It Does
Description here
Cloud Services
Service A, Service BEst. Monthly Cost
$0-10
Why At This Scale
Reason here
Infrastructure as Code — Terraform (Startup)
# AWS ALB + Auto-Scaling Group + RDS
# Production-ready for growing startups
terraform {
required_providers {
aws = {
source = "hashicorp/aws"
version = "~> 5.0"
}
}
}
# Application Load Balancer
resource "aws_lb" "app_alb" {
name = "observatory-alb"
internal = false
load_balancer_type = "application"
security_groups = [aws_security_group.alb.id]
subnets = var.public_subnet_ids
}
# Auto-Scaling Group
resource "aws_autoscaling_group" "app_asg" {
name = "observatory-asg"
min_size = 2
max_size = 10
desired_capacity = 2
health_check_type = "ELB"
launch_template {
id = aws_launch_template.app.id
version = "$Latest"
}
}
# RDS PostgreSQL Instance
resource "aws_db_instance" "primary" {
identifier = "observatory-db"
engine = "postgres"
engine_version = "15.4"
instance_class = "db.t3.medium"
allocated_storage = 100
multi_az = true
storage_encrypted = true
}
Click to Learn More
Line 13-20
Application Load Balancer — Distributes traffic across app servers. Handles SSL termination and health checks.
Line 23-35
Auto-Scaling Group — Maintains 2-10 instances based on CPU/memory metrics. Zero-downtime deployments.
Line 38-48
RDS with Multi-AZ — Automatic failover to standby. Encrypted storage for compliance.
Architecture Diagram — Scale
Component Name
COMPONENT TYPEWhat It Does
Description here
Cloud Services
Service A, Service BEst. Monthly Cost
$0-10
Why At This Scale
Reason here
Infrastructure as Code — Terraform (Scale)
# Kubernetes Deployment + Service + Ingress
# Global scale with multi-region support
terraform {
required_providers {
kubernetes = {
source = "hashicorp/kubernetes"
version = "~> 2.23"
}
aws = {
source = "hashicorp/aws"
version = "~> 5.0"
}
}
}
# EKS Cluster
resource "aws_eks_cluster" "main" {
name = "observatory-cluster"
role_arn = aws_iam_role.cluster.arn
version = "1.28"
vpc_config {
subnet_ids = var.private_subnet_ids
endpoint_private_access = true
endpoint_public_access = true
}
}
# Kubernetes Deployment
resource "kubernetes_deployment" "app" {
metadata {
name = "observatory-app"
}
spec {
replicas = 6
selector {
match_labels = { app = "observatory" }
}
template {
metadata { labels = { app = "observatory" } }
spec {
container {
name = "app"
image = "observatory/app:latest"
resources {
limits = {
cpu = "1000m"
memory = "1Gi"
}
}
}
}
}
}
}
# Ingress with AWS ALB Controller
resource "kubernetes_ingress" "app" {
metadata {
name = "observatory-ingress"
annotations = {
"kubernetes.io/ingress.class" = "alb"
"alb.ingress.kubernetes.io/scheme" = "internet-facing"
}
}
}
Click to Learn More
Line 17-30
EKS Cluster — Managed Kubernetes control plane. Private + public endpoints for security.
Line 33-60
K8s Deployment — 6 replicas with resource limits. Rolling updates with zero downtime.
Line 63-73
Ingress + ALB — AWS ALB Controller provisions load balancer automatically. SSL via ACM.
Cost Comparison Across Scales
Primary Cost Drivers
Hobby / MVP
- Free tier CDN (Cloudflare)
- Serverless compute (pay-per-request)
- Managed DB free tier (Supabase/D1)
- Single region, no redundancy
Startup
- ALB + 2-10 EC2 instances
- RDS with Multi-AZ failover
- ElastiCache Redis cluster
- S3 storage + data transfer
Scale
- EKS cluster + node groups
- Aurora Global Database
- Multi-region deployment
- Redis Cluster + MQ + Workers
- Monitoring & observability stack