GCP Kubernetes Microservices Project

Overview

This project is a microservices architecture deployed on Google Kubernetes Engine (GKE). It consists of two containers, Container 1 and Container 2, which communicate bidirectionally to perform file storage, retrieval, and product calculation tasks. The infrastructure is provisioned using Terraform, and CI/CD pipelines are set up using GCP Cloud Source Repository and Cloud Build.

Architecture Diagram

Container 1

Container 1 serves as the gateway for storing files to a persistent volume in GKE and calculating products from the stored files. It exposes REST API endpoints for file storage and product calculation.

Features

• Store File Endpoint: Accepts JSON input to create and store files in GKE persistent storage.
• Calculate Product Endpoint: Accepts JSON input to calculate the total of a product from the stored file.

JSON Input:

{
  "file": "file.dat",
  "data": "product, amount \nwheat, 10\nwheat, 20\noats, 5"
}

JSON Output:

{
  "file": "file.dat",
  "message": "Success."
}

Example file.dat:

product, amount wheat,10 wheat,20 oats,5

Container 2

Container 2 listens on a specified port and endpoint to interact with Container 1. Its primary role is to calculate the total of a product by summing rows of the same product from the given file.

Features

• Product Calculation: Calculates the total of a product from the given file.

Additional Requirements

• CI/CD Pipeline: Utilizes GCP Cloud Source Repository and Cloud Build for automated building and deployment of code changes.
• Terraform Script: Infrastructure provisioning is automated using Terraform scripts.
• Cluster Configuration: Configures the GKE cluster according to specified requirements to minimize costs.
• Persistent Volume: Attaches a persistent volume to the GKE cluster for file storage.
• Service Exposure: Exposes Container 1 service to the internet for external access.

How to Use

1. Clone the repository.
2. Navigate to the kubernetes-service<ServiceNumber> folder.
3. Deploy the pvc.yaml file for mounting the persisting volume claim of 1Gi.
4. Deploy the Kubernetes workload using the provided YAML file.
5. Access the REST API endpoints of Container 1 for file storage and product calculation.