🧅 Onion Price Tracker – How It Works

Check out the tracker here if you haven't already.

🛠️ Tech Stack

• AWS S3 – Stores static web pages, daily price JSON files, and the city_state.json mapping.
• AWS CloudFront – Serves static assets globally via CDN for fast access.
• AWS Lambda – Runs serverless functions for data ingestion and processing.
• AWS DynamoDB – Stores raw onion price records with city and date.
• AWS EventBridge – Schedules Lambda executions automatically.

---

Architecture

📂 Data Flow

1. Data Ingestion

• A Lambda function fetches onion prices from the Government API.
• Prices are averaged at the city level.
• Data is stored in DynamoDB with schema:
  • city: Partition key
  • date: Sort key
  • price

---

2. Data Processing

• Another Lambda function generates JSON files (city_date.json) by reading from DynamoDB data.
• This function is configured as output of the ingestion lambda so that it is triggered after it.
• Each file contains data for the last 6 months for a city
• These JSON files are written to an S3 bucket.

---

3. Data Serving

• The webpage fetches JSON files from S3.
• CloudFront serves both:
  • Static HTML/CSS/JS files.
  • City-specific JSON data files.

---

4. Scheduling

• EventBridge triggers the Lambda functions once daily.

---

⚡ Additional Details

• Entire system runs within the AWS Free Tier (Lambda free invocations, DynamoDB free reads/writes, S3 free storage/requests).
• Data retention is capped at 6 months per city (older entries expire automatically via TTL).
• The list of supported cities is stored in a static JSON (city_state.json) in S3, making the system configurable without code changes.

---

✅ This pipeline ensures that visitors always see the latest onion prices per city, fetched once per day, stored for 6 months, and served globally through S3 + CloudFront — all while staying within AWS Free Tier.