Distributed Collaborative Paint Canvas 🎨

A real-time, distributed collaborative drawing application designed to demonstrate the principles of Distributed Systems, High Availability, and GitOps-driven Cloud Orchestration.

---

🏗 Choose Your Architecture

This project evolved from a single-node VPS deployment into a professional, multi-AZ AWS cloud infrastructure. You can deploy it in two ways:

1. **AWS Production Deployment **

• Infrastructure: AWS EKS (Kubernetes), RDS (PostgreSQL), ElastiCache (Redis), Secrets Manager, ALB.
• Logic: Fully automated via Terraform (IaC) and GitHub Actions (CI/CD).

2. **Local / VPS Deployment **

• Infrastructure: Lightweight k3s cluster, Docker Compose, or manual Node.js.
• Logic: Standard kubectl apply manifests.
• Location: Legacy manifests are preserved in the k83-vps/ directory.

---

🚀 AWS Enterprise Deployment (EKS)

Prerequisites

• AWS CLI configured with Admin permissions.
• Terraform 1.5+
• kubectl & Helm

Step 1: Bootstrap the State

Before deploying the main resources, initialize the remote S3 backend and DynamoDB state lock:

cd terraform/bootstrap
terraform init && terraform apply

Step 2: Provision Infrastructure

Update terraform/main/provider.tf with the S3 bucket name from the bootstrap step, then:

cd terraform/main
terraform init
terraform apply

This will provision a custom VPC, EKS Cluster, RDS Postgres, and ElastiCache Redis (~15-20 mins).

Step 3: GitOps Rollout

1. Add your AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY to GitHub Repository Secrets.
2. Push a change to the master branch.
3. The Deployment Workflow will automatically build the images, push to ECR, and deploy to EKS.

---

🐧 Legacy VPS Deployment (k3s)

If you are deploying to a standard Linux VPS using k3s:

1. Ensure k3s is installed: curl -sfL https://get.k3s.io | sh -
2. Deploy Manifests:

   kubectl apply -f k83-vps/

3. Config: Update the Ingress in k83-vps/ingress.yaml with your VPS IP or public domain.

---

🛠 Tech Stack

Component	AWS Technology	VPS/Local Technology
Kubernetes	Amazon EKS (Managed)	k3s (Lightweight)
Database	Amazon RDS (PostgreSQL)	In-cluster Postgres Pod
Cache	Amazon ElastiCache (Redis)	In-cluster Redis Pod
Ingress	AWS Load Balancer (ALB)	Traefik / Nginx
Secrets	AWS Secrets Manager	K8s Secrets (Base64)
Registry	Amazon ECR	Docker Hub
IaC	Terraform	Manual Manifests

---

📉 Stress Testing & HPA

The system is built to breathe. We’ve included a "Bot Army" tester to verify Horizontal Pod Autoscaling (HPA).

1. Install Tester:

   cd tester
   npm install

2. Run "Starry Night" Attack:

   # Simulates 250 concurrent bots drawing a Van Gogh painting
   CONCURRENCY=250 INTERVAL=10 node flood.js

3. Monitor Scaling:

   kubectl get hpa paint-backend -w

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.