Serverless Authentication System with AWS & JWT

This project implements a secure, scalable serverless authentication system using AWS services including API Gateway, Lambda, and DynamoDB, with JWT (JSON Web Token) for session management.

Architecture Overview

The system implements a modern serverless approach with two primary workflows:

User Authentication Flow

• Client application initiates the authentication process
• System checks if user is registered
• New users follow registration path
• Existing users proceed to login
• Login credentials are verified against DynamoDB
• JWT tokens are issued for successful authentications

API Management Flow

• Client applications communicate with backend services through API Gateway
• API Gateway manages and secures all API requests
• Lambda functions (written in Go) handle business logic
• DynamoDB provides scalable NoSQL data storage for user data

Key Features

• Serverless Architecture: Eliminates server management using AWS Lambda and managed services
• Secure Authentication: Implements industry-standard JWT authentication
• Scalable Design: Leverages AWS auto-scaling capabilities for handling variable loads
• Cost-Effective: Pay-only-for-usage pricing model
• Go Implementation: Backend services written in efficient Go code
• API-Driven Design: RESTful endpoints for all authentication operations

Technology Stack

The system comprises the following core AWS components:

• AWS Lambda: Serverless compute for authentication logic
• Amazon API Gateway: API management and security
• Amazon DynamoDB: NoSQL database for user credentials and application data
• JWT (JSON Web Tokens): Secure authentication tokens
• Go (Golang): Backend programming language
• AWS IAM: Identity and Access Management for secure service interactions

Implementation Details

The authentication flow works as follows:

1. Registration Process:

   • New users submit registration details through client application
   • API Gateway routes request to Lambda function
   • Lambda validates input and stores user data in DynamoDB
   • Success/failure response returned to user

2. Login Process:

   • Users submit credentials through client application
   • API Gateway routes request to authentication Lambda
   • Lambda verifies credentials against DynamoDB
   • On success, JWT token is generated and returned
   • On failure, error message is returned

3. Protected API Access:

   • Client includes JWT in authorization header
   • API Gateway validates token before processing request
   • Valid tokens allow access to protected resources
   • Invalid tokens return authorization errors

Deployment & Testing

The project uses AWS CDK for infrastructure deployment and is managed through a Makefile for simplified operations.

Available Deployment & Testing Commands

Available commands:
  make diff         - Show differences between local and deployed stacks
  make deploy       - Deploy CDK stack(s) (runs diff first)
  make destroy      - Destroy CDK stack(s) with confirmation prompt
  make clean        - Remove local CDK-generated files
  make all          - Run diff and deploy
  make help         - Show this help message
  make test-api     - Test deployed apis, and check integrity

To specify a stack, use STACK_NAME=your-stack-name
Example: make deploy STACK_NAME=my-service-stack