🌌 CosmicLoad: GCP Global Load Balancer

🌟 Global LoadBalancer Response

🌟 Project Overview

CosmicLoad is a cloud infrastructure project designed to explore and implement 🌐 Google Cloud Platform's (GCP) Global Load Balancer. This project demonstrates how to effectively distribute user requests across multiple regions and availability zones, ensuring:

• ✅ High Availability
• 🚀 Fault Tolerance
• ⚡️ Minimal Latency
  by leveraging Google's global infrastructure.

🔥 Key Features

1. Global Load Balancing 🌎: Distributes traffic between two regions (US Central & Asia Southeast).
2. Instance Templates 🖥️: Built with Ubuntu OS.
3. Instance Groups 🛠️:
   • us-central (Iowa): Includes 4 availability zones.
   • asia-southeast1 (Singapore): Includes 3 availability zones.
4. Cloud Armour 🛡️: Blocks requests originating from India to improve security and control access.(Due to free-account gcp restricts cloud-armour quota)
5. Auto Scaling 📈: Dynamically scales instances based on CPU utilization:
   • us-central: Min 4 instances, max 8 instances.
   • asia-southeast1: Min 3 instances, max 6 instances.
6. Traffic Distribution 🔄: Uses the global load balancer to distribute traffic and ensure low-latency access to users worldwide.

💻 Components Breakdown

1. Instance Templates 🖥️

• Configured GCE Ubuntu instances running Apache web server in:
  • us-central-a (Iowa)
  • asia-southeast1-a (Singapore)

2. Instance Groups 🚀

• us-central: Autoscaled group across 4 availability zones (a, b, c, f) with 4 minimum and 8 maximum instances.
• asia-southeast1: Autoscaled group across 2 availability zones (a, b) with 3 minimum and 6 maximum instances.

3. Global Load Balancer 🌐

• Configured a Global Load Balancer to distribute incoming traffic across the instance groups in both regions. This ensures:
  • 📊 High Availability: Continues serving traffic even if a region fails.
  • ⚡️ Low Latency: Directs users to the region closest to them for faster response times.

4. Cloud Armour 🛡️

• Implemented Cloud Armour to block traffic from India. This helps:
  • 🛠️ Filter out requests from unwanted regions.
  • 🧑‍💻 Maintain security by controlling access based on geographical locations.

5. Auto-Scaling 📈

• Autoscaling based on CPU utilization to optimize resource usage and reduce costs:
  • us-central: Scales between 4 to 8 instances.
  • asia-southeast1: Scales between 3 to 6 instances.

---

🔨 Steps to Reproduce

Prerequisites ✅

1. GCP Account with billing enabled.
2. Knowledge of GCP services like Compute Engine, Load Balancer, and Cloud Armour.
3. Optional: Familiarity with Terraform or gcloud CLI for resource management.

🛠️ 1. Create Instance Templates

• Use the GCE Ubuntu image and install the Apache web server on the instances in both regions.

🛠️ 2. Create Instance Groups

• Create an instance group for each region:
  • us-central with autoscaling (min 4, max 8).
  • asia-southeast1 with autoscaling (min 3, max 6).

🛠️ 3. Set Up Global Load Balancer 🌎

• Configure a GCP Global Load Balancer to route incoming traffic across the instance groups in both regions.

🛠️ 4. Configure Cloud Armour 🛡️

• Set up Cloud Armour to block requests from India or any other unwanted geographic location.

🛠️ 5. Test and Verify ⚙️

• Send user traffic from different locations to verify:
  • ✅ Traffic distribution across regions.
  • 🚫 Requests from blocked regions are denied.
  • 📈 Instances are auto-scaling based on load.

---

🧰 Tech Stack

• Google Cloud Platform (GCP):
  • 🌍 Compute Engine
  • ⚡️ Global Load Balancer
  • 🔐 Cloud Armour
• Ubuntu: Operating system on the instances.
• Apache Web Server: Hosting web services on each instance.
• Terraform: Optional, for automating infrastructure provisioning.