How I Provisioned Scalable AWS Infrastructure with Terraform and Load Balancer
Kaustav Dey
canonical_url: https://github.com/KaustavDey357/Terraform-AWS-EC2-Load-Balancer-Deployment
As I deepened my understanding of DevOps and cloud fundamentals, I wanted to get hands-on with provisioning infrastructure the right way: using Infrastructure as Code (IaC). In this post, I’ll walk through how I built a reusable Terraform project to provision an EC2 instance, attach a load balancer, configure security groups, and get it all running on AWS.
This project was part of my journey in creating modular and production-oriented DevOps blueprints.
🧰 Tools & Tech
Terraform for IaC
AWS EC2 for compute resources
Application Load Balancer (ALB) for routing
Security Groups for access control
Remote backend support with Terraform state files
📦 Project Structure
I broke the configuration into reusable, modular Terraform components to make the codebase scalable and production-ready. Here's how the structure looked:
.
├── main.tf # Orchestrator: may call modules or glue everything together
├── variables.tf # All variable declarations with types and descriptions
├── outputs.tf # Output values (e.g., IPs, DNS names, ARNs)
├── vpc.tf # VPC, subnets, internet gateway, etc.
├── ec2.tf # EC2 instance(s), AMIs, key pairs, EBS volumes
├── alb.tf # Application Load Balancer, listeners, target groups
├── security.tf # Security groups, network ACLs, firewall rules
├── README.md # Project documentation
├── .gitignore # Files to exclude from Git (e.g., `.terraform`, `*.tfstate`)
Each module encapsulates a piece of the infrastructure (e.g., EC2, security group), keeping things clean and reusable.
🏗️ What It Provisions
When executed, the Terraform code provisions:
A public subnet in a selected region
An EC2 instance with user data for bootstrapping
A Security Group that allows inbound traffic on ports 22 and 80
An Application Load Balancer (ALB) that distributes HTTP traffic
Target group + listener configuration for the EC2 instance
🧪 How to Use It
Clone the repo and run the following inside the root directory:
terraform init
terraform plan
terraform apply
Make sure your AWS credentials are set in your environment or shared credentials file.
Once applied, the EC2 instance and ALB will be up and running. You’ll get the public DNS of the load balancer in the Terraform output.
✅ Outcome
With a single command, I spun up a complete production-grade architecture using Terraform. It’s scalable, reusable, and can easily be extended to include databases, autoscaling, and monitoring.
📌 Key Learnings
Writing modular Terraform code is essential for maintainability
ALBs are ideal for HTTP/HTTPS workloads with flexible routing
Outputs and variables improve reusability and flexibility
Infrastructure automation saves time and reduces errors
🧠 Next Steps
I’m planning to:
Add support for private subnets and NAT gateways
Integrate with RDS or DynamoDB
Add Terraform Cloud remote backend
Extend this into a full production deployment pipeline
🔗 Resources
- GitHub Repo: Terraform AWS EC2 Load Balancer Deployment
☎️ Let's Connect
If you're building something cloud-native or want help setting up secure AWS infrastructure, I’d love to chat.
Subscribe to my newsletter
Read articles from Kaustav Dey directly inside your inbox. Subscribe to the newsletter, and don't miss out.
Written by