Real World Application with DynamoDB and Single-Table Design (We Work Remotely Edition)

Are you looking to get the most out of DynamoDB? Look no further!

We're going to design a high performance database with DynamoDB and Single-Table Design for a real world application - We Work Remotely. We Work Remotely is the largest remote work community in the world with over 4.5M visitors.

In this tutorial, we will:

1. Design entity relationship diagram (ERD)
2. Identify the access patterns
3. Walk through single-table design data modelling process
4. Cost estimation based on 1.7M monthly visitors

Table of Contents:

• TL;DR
• Introduction
• ERD & Access Patterns
• Data Modelling Walkthrough
  • Modelling Draft, Featured & normal Job Post Entity
    1. Fetch single job post (applied to Featured & Draft) record
    2. Fetch featured & latest job posts for company
    3. Fetch featured & latest job posts for category
    4. Fetch draft job posts for company (Optional)
    5. Promote draft to featured/normal job post
    6. Downgrade featured job post to normal job post after expiry (7 days)
    7. Take down expired job posts from public after 30 days
    8. Fetch draft & expired job posts for company (Optional)
  • Handling "Apply for this position"
  • Send New Job Posts to Subscribers Daily
  • View all Companies
    1. Static partition ("COMPANIES") on GSI
    2. Sparse index (CompanyIndex) to project only Companies entity
• Conclusion
• Cost Calculation
  • Traffic Estimation
  • Item Size Estimation
  • WCU/RCU Estimation