🏗️ AWS 3-Tier Architecture with Terraform

📚 Overview

This Terraform project implements a 3-Tier Architecture on AWS, dividing the application infrastructure into three logical layers:

• Presentation Layer (Public Subnet + ALB)
• Application Layer (EC2 Instances in Private Subnets with Auto Scaling)
• Data Layer (Amazon RDS in Private Subnets)

The infrastructure is deployed in a custom VPC, spans two Availability Zones, and uses modular Terraform design for scalability, reusability, and best practices.

---

🧱 Architecture Diagram

---

🔧 Terraform Modules Breakdown

1️⃣ Networking Module

Sets up the base networking layer, ensuring secure and isolated deployment.

Resources Created:

• aws_vpc: Virtual Private Cloud
• aws_subnet: Public and Private subnets (across AZ1 and AZ2)
• aws_internet_gateway: Internet access for public subnet
• aws_nat_gateway: Allows private subnet instances to reach the internet
• aws_route_table and aws_route_table_association: Routing rules

📌 Subnet Usage:

• Public Subnet: For Load Balancer, NAT Gateway
• Private Subnet: For EC2 (App Layer) and RDS (Data Layer)

---

2️⃣ Application Load Balancer (ALB) Module

Serves as the public-facing entry point for web traffic.

Resources Created:

• aws_lb: Application Load Balancer
• aws_lb_listener: For HTTP/HTTPS ports (80/443)
• aws_lb_target_group: Routes requests to EC2 instances
• aws_security_group: Controls access to the ALB

✅ ALB ensures traffic routing, health checks, and scaling integration.

---

3️⃣ Auto Scaling Group (ASG) Module

Provisions and manages EC2 instances in the Application Layer.

Resources Created:

• aws_launch_template: EC2 config (AMI, instance type, user data)
• aws_autoscaling_group: Automatically scales instances
• aws_cloudwatch_metric_alarm: CPU or custom metric-based alarms
• aws_autoscaling_policy: Scaling rules (e.g., CPU > 70%)

🚀 EC2 instances are deployed in private subnets and are registered with the ALB target group.

---

4️⃣ Database (RDS) Module

Deploys a secure and fault-tolerant Data Layer using Amazon RDS.

Resources Created:

• aws_db_subnet_group: Deploys DB across private subnets in multiple AZs
• aws_db_instance: Managed relational DB (e.g., MySQL/PostgreSQL)

🔒 No public access is provided; only EC2s from the Application Layer can access RDS.

---

5️⃣ Root Module: Integration

The root module orchestrates all other modules and wires the dependencies via Terraform inputs and outputs.

Example Integrations:

• alb module receives vpc_id and public_subnet_ids from networking
• asg module receives target_group_arn from alb, and subnet_ids from networking
• database module receives private_subnet_ids from networking

---

🔄 Request Flow

1. User sends request via the internet.
2. ALB in public subnet receives request.
3. Request forwarded to EC2 instance in private subnet.
4. EC2 processes request and, if needed, queries RDS in another private subnet.
5. Response flows back via ALB.

---

💡 Benefits of 3-Tier Architecture

• 🔒 Security: Private subnets, IAM roles, SGs, and no public DB access
• 📈 Scalability: Auto Scaling and Load Balancing
• 🧩 Modularity: Layered infrastructure via reusable modules
• 🔁 High Availability: Multi-AZ RDS and distributed subnets

---

📁 Folder Structure (Example)

.
├── modules/
│   ├── networking/
│   ├── alb/
│   ├── asg/
│   └── database/
├── main.tf
├── variables.tf
├── outputs.tf
└── README.md