C100DEV Data Modeling Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Embed all child documents inside the parent document.

  Why it's wrong here

  Embedding a 'squillion' items will instantly violate the 16MB document size limit. This approach is only viable for small, bounded relationships. For massive datasets, this pattern leads to application failure and data loss, making it the worst possible choice for this specific scenario.

• ✗

  Store an array of child IDs in the parent document.

  Why it's wrong here

  Storing an array of references in the parent still leads to an unbounded array growth issue. As the number of children grows, the parent document will eventually exceed the 16MB limit, and updating the array on every child insertion will cause significant performance bottlenecks and write contention.

• ✓

  Store the parent ID in the child documents.

  Why this is correct

  Storing the parent ID in each child document is the standard 'referencing' approach for massive relationships. This keeps both parent and child documents small, avoids size limits, and allows you to index the parent ID on the child collection for fast, scalable retrieval of all related items.

• ✗

  Use a separate collection for each child to ensure isolation.

  Why it's wrong here

  Creating a separate collection for each child is a significant anti-pattern that leads to unmanageable schema sprawl. It makes global queries across children impossible and destroys the ability to perform efficient indexing or aggregation. This approach would make the system essentially impossible to maintain or scale effectively over time.