Text Message Marketing Campaign Micro-Service

Overview

This project implements a text message marketing campaign micro-service designed to handle high-throughput text message sending with a rate limit of 45 segments per second. It includes robust error handling, rate limiting, and logging mechanisms.

Features

Message Queuing: Handles incoming text message requests, calculates segments, and queues them for processing.
Rate Limiting: Ensures the sending rate does not exceed 45 segments per second.
Error Handling: Implements retry logic and logs errors and successes.
API Gateway: Exposes the QueueMessage function as a RESTful endpoint.

Project Structure

|-- packages/
|   |-- functions/
|       |-- src/
|           |-- utils.ts
|           |-- message.ts
|           |-- queueMessage.ts
|       |-- test/
|           |-- utils.specs.ts
|-- README.md
|-- .gitignore
|-- package.json
|-- tsconfig.json
|-- packages/
|   |-- stacks/
|           |-- MyStacks.ts
|-- sst.config.ts

Prerequisites

Node.js (version 14 or higher)
AWS Account (for deploying AWS resources such as Lambda, SQS, DynamoDB)
TypeScript
SST (Serverless Stack)

Setup Instructions

1. Clone the Repository

git clone git@github.com:kalib-code/coding-challenge-2024.git
cd coding-challenge-2024

2. Install Dependencies

npm install

3. Configure Environment Variables

Everything is configured in the sst.config.ts file.

4. Set Up AWS Resources

Make use you install aws cli and configure your credentials. docs.aws.amazon.com/cli/latest/userguide/cli-configure-quickstart.html

5. Deploy the Service (using Serverless Framework as an example)

npx sst deploy

6. Run Tests

npm sst test

Usage

API Endpoints

QueueMessage

Endpoint: POST /send Description: Queues new text message requests.

Payload:

{
    "message": "Your SMS content here"
}

Response:

200 OK: Messages queued successfully.
400 Bad Request: Validation error or invalid payload.
500 Internal Server Error: Server error.

Example Request

curl -X POST https://your-api-endpoint/send \
     -H "Content-Type: application/json" \
     -d '{"message": "Hello, this is a test message."}'

Code Overview

`src/utils.ts`

Contains utility functions:

canSendSegments(): Implements a leaky bucket algorithm to enforce rate limiting.
getBucketState() and updateBucketState(): Manage leaky bucket state using DynamoDB.

`src/message.ts`

Contains the processMessage function for sending segments while respecting the rate limit.

`src/queueMessage.ts`

Contains the Lambda handler for queuing messages, validating input, calculating segments, and sending chunks to SQS.

`test/utils.specs.ts`

Contains test placeholders for unit testing the utility functions.

Error Handling

Implements retry logic with exponential backoff for handling transient errors.
Logs errors and successes to the console (consider using AWS CloudWatch for production).

Scalability and Performance

instead of dynamoDB we can use redis for storing the bucket state as it is faster and more scalable.
using Dead Letter Queues (DLQ) for handling failed messages. This allows us to reprocess failed messages and ensure no message is lost.
using AWS Lambda Provisioned Concurrency to reduce cold starts and improve performance.
adding cloudwatch alarms to monitor the service and set up autoscaling based on the number of messages in the queue.

kalib-code / coding-challenge-2024