Heath Cloud Native

This project is a cloud-native application deployed on Google Cloud Platform using Terraform. The application source code is taken from this repo. The cloud deployment architecture and infrastructure automation (Terraform, Kubernetes, Cloud Functions, Load Testing) are developed for the CmpE 48A Cloud Computing course by:

• Kerem Haktan Kurt
• Mustafa Batuhan Büber

Project Structure

• heatHBack: Spring Boot Backend Application.
• heatHFront: React Frontend Application.
• terraform: Infrastructure as Code (IaC) configuration to provision GCP resources.
• k8s: Kubernetes manifests.
• functions: Google Cloud Functions (Java).
• locust: Load testing scripts using Locust.

Prerequisites

Ensure you have the following tools installed on your local machine:

• Google Cloud SDK (gcloud)
• Terraform (v1.5.0+)
• kubectl (optional, for cluster management)

Setup & Deployment Instructions

1. Google Cloud Authentication

First, authenticate with Google Cloud to allow Terraform to create resources.

1. Login to your Google Cloud account:

   gcloud auth login

2. Set your project ID:

   gcloud config set project <your-gcp-project-id>

3. Check your project ID:

   gcloud config get-value project

4. Grant Cloud Build permissions:

   gcloud projects add-iam-policy-binding <your-gcp-project-id> --member="user:<your-email>" --role="roles/cloudbuild.builds.editor"

5. Configure Application Default Credentials (ADC) for Terraform:

   gcloud auth application-default set-quota-project <your-gcp-project-id>

2. Configure Terraform

1. Navigate to the terraform directory:

   cd terraform

2. Open terraform.tfvars and update the values for your environment.

   • Important: You must update the project_id to match your GCP Project ID.
   • Update sendgrid_api_key and sender_email if you want email functionality to work.

   Example terraform.tfvars:

   project_id = "your-gcp-project-id"

3. Deploy Infrastructure

1. Initialize Terraform:

   terraform init

2. Plan the deployment to see what will be created:

   terraform plan

3. Apply the configuration to create the infrastructure:

   terraform apply

   • Type yes when prompted to confirm.
   • Note: This process automatically builds the Docker images for the Frontend and Backend using Google Cloud Build and pushes them to the Artifact Registry. The initial deployment can take 15-20 minutes.

4. Access the Application

Once terraform apply completes successfully, you will see several outputs in your terminal:

• frontend_external_ip: The public IP address to access the web application. Open this in your browser.
• vm_public_ip: The public IP of the PostgreSQL Database VM.
• get_credentials_command: A command to configure kubectl to connect to your new GKE cluster.

To view these outputs again later, run:

terraform output

Testing & Monitoring

• Kubernetes: Use the get_credentials_command output to configure kubectl, then use standard commands like kubectl get pods to check the status of your deployments.

• Load Testing with Locust: The locust directory contains scripts for load testing the application's auto-scaling capabilities.

  1. Install Locust:

     pip install locust

  2. Configure Test Parameters:

     • Load Config: You can modify USERS and SPAWN_RATE directly in locust/run_hpa.py to simulate different traffic patterns.
     • Infrastructure Config: You can tune scaling behavior (like gke_max_node_count, backend_hpa_max_replicas) in terraform/terraform.tfvars and apply changes with terraform apply.

  3. Run the Test:

     python locust/run_hpa.py

     This will start the load test and generate HTML reports in the locust/results_hpa directory, visualizing HPA scaling, Pod CPU usage, and Database Connection Pool status.

Cleanup

To destroy all resources and avoid incurring further costs:

cd terraform
terraform destroy