Overview

This project demonstrates how to build a scalable, resilient blog application deployed across AWS, GCP, and Azure simultaneously. Each cloud provider hosts a different part of the architecture to maximize availability and avoid vendor lock-in.

Multicloud Architecture

Main components:

• Frontend (GCP): Cloud Run serving the web app container
• Database (AWS): Amazon RDS PostgreSQL for structured blog data
• Media Storage (Azure): Blob Storage for post images and uploads
• Global CDN: Cloudflare or CloudFront to unify multi-cloud endpoints

                    +------------------+
                    |   Cloudflare     |
                    | Global CDN & DNS |
                    +--------+---------+
                             |
             --------------------------------------------------
             |                       |                        |
     +-------v-------+       +-------v-------+        +-------v-------+
     |     GCP       |       |      AWS       |       |     Azure     |
     | Cloud Run     |       | RDS PostgreSQL |       | Blob Storage  |
     | Frontend API  |<----->| Main Database  |<----->| Media Files   |
     +---------------+       +---------------+        +---------------+
        

Why Multicloud?

• High resilience — if one provider fails, others stay online
• Avoid vendor lock-in
• Use the best tool from each cloud
• Latency optimization with global routing

1. Deploy the Frontend to GCP Cloud Run

Enable required services

gcloud services enable run cloudbuild.googleapis.com artifactregistry.googleapis.com

Dockerfile

FROM python:3.10
WORKDIR /app
COPY . .
RUN pip install -r requirements.txt
CMD ["gunicorn", "-b", ":8080", "app:app"]

Build & Deploy

gcloud builds submit --tag gcr.io/multicloud-proj/frontend
gcloud run deploy blog-frontend \
  --image gcr.io/multicloud-proj/frontend \
  --allow-unauthenticated \
  --region us-central1 \
  --set-env-vars DB_HOST=<AWS_RDS_HOST>,AZURE_CONTAINER_URL=<URL>

2. Set Up the Database on AWS RDS (PostgreSQL)

Create RDS Instance

aws rds create-db-instance \
  --db-instance-identifier blogdb \
  --db-instance-class db.t3.micro \
  --engine postgres \
  --master-username admin \
  --master-user-password MySecurePass123 \
  --allocated-storage 20

After a few minutes, retrieve the RDS endpoint:

aws rds describe-db-instances --query "DBInstances[*].Endpoint.Address"

3. Create the Media Bucket on Azure Blob Storage

Create Resource Group

az group create -n blog-multicloud -l westeurope

Create Storage Account

az storage account create \
  -n blogstorage123 \
  -g blog-multicloud \
  -l westeurope \
  --sku Standard_LRS

Create Container

az storage container create \
  -n images \
  --account-name blogstorage123

4. Application Configuration

The Flask backend running on Cloud Run uses:

DB_HOST=<aws-rds-endpoint>
AZURE_CONTAINER_URL=https://blogstorage123.blob.core.windows.net/images

Flask Example

image_url = f"{os.getenv('AZURE_CONTAINER_URL')}/{filename}"

conn_string = f"postgresql://admin:MySecurePass123@{os.getenv('DB_HOST')}:5432/blogdb"
        

5. Integrate a Global Load Balancer

Recommended options:

• Cloudflare – easiest to configure and cloud-agnostic
• AWS CloudFront → origin GCP Cloud Run

Cloudflare
  DNS → CNAME → Cloud Run URL
  Page Rule → Cache static assets
  Security → WAF + DDoS protection

6. CI/CD Across Clouds

Use GitHub Actions as the unified CI/CD orchestrator.

Example GitHub Workflow

name: Multicloud Deploy

on:
  push:
    branches: [ "main" ]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
    - uses: actions/checkout@v3

    - name: Deploy to GCP Cloud Run
      run: |
        gcloud builds submit --tag gcr.io/$PROJECT/frontend
        gcloud run deploy blog-frontend --image gcr.io/$PROJECT/frontend

    - name: Sync Media to Azure
      run: |
        az storage blob upload-batch -d images -s static/

Advantages & Disadvantages of Multicloud

Advantages	Disadvantages
High availability and improved resilience	More complex to configure and maintain
Avoid vendor lock-in	Networking between clouds introduces latency
Use the best service from each provider	Costs harder to predict
Better global performance with multi-region options	Security policies must be synchronized everywhere